This week, the Trump Administration proposed a new policy framework for artificial intelligence (AI) technologies that attempts to balance the need for continued innovation with a set of principles to address concerns about new AI services and applications. This represents an important moment in the history of emerging technology governance as it creates a policy vision for AI that is generally consistent with earlier innovation governance frameworks established by previous administrations.

Generally speaking, the Trump governance vision for AI encourages regulatory humility and patience in the face of an uncertain technological future. However, the framework also endorses a combination of “hard” and “soft” law mechanisms to address policy concerns that have already been raised about developing or predicted AI innovations.

AI promises to revolutionize almost every sector of the economy and can potentially benefit our lives in numerous ways. But AI applications also raise a number of policy concerns, specifically regarding safety or fairness. On the safety front, for example, some are concerned about the AI systems that control drones, driverless cars, robots, and other autonomous systems. When it comes to fairness considerations, critics worry about “bias” in algorithmic systems that could deny people jobs, loans, or health care, among other things.

These concerns deserve serious consideration and some level of policy guidance or else the public may never come to trust AI systems, especially if the worst of those fears materialize as AI technologies spread. But how policy is formulated and imposed matters profoundly. A heavy-handed, top-down regulatory regime could undermine AI’s potential to improve lives and strengthen the economy. Accordingly, a flexible governance framework is needed and the administration’s new guidelines for AI regulation do a reasonably good job striking that balance.

Background

Last February, the White House issued Executive Order 13859, on “Maintaining American Leadership in Artificial Intelligence.” The Order announced the creation of the “American AI Initiative,” an effort to “focus the resources of the Federal government to develop AI.” It prioritized investments in AI-focused research and development (R&D), building a workforce ready for the AI era, international engagement on AI priorities, and the establishment governance standards for AI systems to “help Federal regulatory agencies develop and maintain approaches for the safe and trustworthy creation and adoption of new AI technologies.”

Regarding that last objective, Order 13589 required the Office of Management and Budget (OMB) and the Office of Science and Technology Policy (OSTP) to develop a framework and set of principles for federal agencies to follow when considering the development of regulatory and non‑regulatory approaches for AI. Importantly, the Order also specified that the framework should seek to “advance American innovation” and “reduce barriers to the use of AI technologies in order to promote their innovative application while protecting civil liberties, privacy, American values, and United States economic and national security.”

That resulted in the memorandum sent to heads of federal departments and agencies this week entitled, “Guidance for Regulation of Artificial Intelligence Applications” (hereinafter AI Guidance). The draft version of the AI Guidance specifies that “federal agencies must avoid regulatory or non-regulatory actions that needlessly hamper AI innovation and growth.” More specifically:

“Agencies must avoid a precautionary approach that holds AI systems to such an impossibly high standard that society cannot enjoy their benefits. Where AI entails risk, agencies should consider the potential benefits and costs of employing AI, when compared to the systems AI has been designed to complement or replace.”

But the AI Guidance is certainly not a call for comprehensive deregulation or the abandonment of all AI federal oversight. The memorandum’s very title reflects an understanding that existing laws and agency rules will continue to play a role in guiding the development of AI, machine-learning, and autonomous systems.

Accordingly, and consistent with past executive orders and OMB regulatory guidance documents for federal agencies, the AI Guidance establishes a set of ten principles that agencies must take into consideration when considering AI policy:

1. Public trust in AI: Requiring that “the government’s regulatory and non-regulatory approaches to AI promote reliable, robust, and trustworthy AI applications, which will contribute to public trust in AI.”
2. Public participation: Agencies must provide “ample opportunities for the public to provide information and participate in all stages of the rulemaking process.”
3. Scientific integrity and information quality: Agencies should “leverage scientific and technical information and processes” to build trust and ensure data quality and transparency.
4. Risk assessment and management: Acknowledging that “all activities involve tradeoffs,” the AI Guidance requires that “a risk-based approach should be used to determine which risks are acceptable and which risks present the possibility of unacceptable harm, or harm that has expected costs greater than expected benefits.”
5. Benefits and costs: As part of those risk assessments, agencies must “carefully consider the full societal costs, benefits, and distributional effects before considering regulations related to the development and deployment of AI applications. Such consideration will include the potential benefits and costs of employing AI, when compared to the systems AI has been designed to complement or replace, whether implementing AI will change the type of errors created by the system, as well as comparison to the degree of risk tolerated in other existing ones.”
6. Flexibility: OMB encourages agencies to “pursue performance-based and flexible approaches that can adapt to rapid changes and updates to AI applications.”
7. Fairness and non-discrimination: Acknowledging that “in some instances, introduce real-world bias that produces discriminatory outcomes or decisions that undermine public trust and confidence in AI,” the AI Guidance requires agencies to consider “issues of fairness and non-discrimination with respect to outcomes and decisions produced by the AI application at issue.”
8. Disclosure and transparency: Agencies are encouraged to consider how greater “transparency and disclosure can increase public trust and confidence in AI applications.”
9. Safety and security: Agencies are required to “promote the development of AI systems that are safe, secure, and operate as intended, and encourage the consideration of safety and security issues throughout the AI design, development, deployment, and operation process.”
10. Interagency coordination: The guidance makes it clear that a “coherent and whole-of-government approach to AI oversight requires interagency coordination.”

Soft Law Ascends

Importantly, the AI Guidance also encourages agencies to be open to “non-regulatory approaches to AI” governance and specifies three particular models:

• Sector-specific policy guidance or frameworks: OSTP writes that “agencies should consider using any existing statutory authority to issue non-regulatory policy statements, guidance, or testing and deployment frameworks, as a means of encouraging AI innovation in that sector.” The memorandum also notes that this can include “work done in collaboration with industry, such as development of playbooks and voluntary incentive frameworks.”
• Pilot programs and experiments: The document encourages the use of “pilot programs that provide safe harbors for specific AI applications” which “could produce useful data to inform future rulemaking and non-regulatory approaches.”
• Voluntary consensus standards: Before regulating, the AI Guidance encourages agencies to consider how voluntary consensus standards, assessment programs, and compliance programs might be used to address policy concerns.

These represent “soft law” approaches to technological governance and they are becoming all the rage in technology policy discussions today. Soft law mechanisms are informal, collaborative, and constantly evolving governance efforts. While not formerly binding like “hard law” rules and regulations, soft law efforts nonetheless create a set of expectations about sensible development and use of technologies. Soft law can include multistakeholder initiatives, best practices and standards, agency workshops and guidance documents, educational efforts, and much more.

Soft law has become the dominant governance approach for emerging technologies because it is often better able to address the “pacing problem,” which refers to the growing gap between the rate of technological innovation and policymakers’ ability to keep up with it. As I have previously noted, the pacing problem is “becoming the great equalizer in debates over technological governance because it forces governments to rethink their approach to the regulation of many sectors and technologies.”

Not only do traditional legislative and regulatory hard law systems struggle to keep up with fast-paced technological changes, but oftentimes those older mechanisms are just too rigid and unsuited for new sectors and developments. That is definitely the case for AI, which is multi-dimensional in nature and even defies easy definition. Soft law offers a more flexible, adaptive approach to learning on the fly and cobbling together principles and policies that can address new policy concerns as they develop in specific contexts, without derailing potentially important innovations.

Building on Past Governance Frameworks

In this sense, the Trump administration’s AI Guidance borrows from past policy frameworks by marrying up a desire to promote an exciting new set of emerging technologies alongside the need for reasonable but flexible oversight and governance mechanisms. At a high level, the AI Guidance builds on many of the same principles that motivated the Clinton administration’s Framework for Global Electronic Commerce, a statement of principles and policy objectives for the then-emerging Internet. The document, which was issued in July 1997, said that “governments should encourage industry self-regulation and private sector leadership where possible” and “avoid undue restrictions on electronic commerce.”

The Framework was a clean break from the top-down regulatory paradigm that had previously governed traditional communications and media technologies. Clinton’s Framework insisted that, to the extent government intervention was needed at all, “its aim should be to support and enforce a predictable, minimalist, consistent and simple legal environment for commerce.” The use of soft law and multistakeholder models was a key component of this vision, and those more flexible governance approaches were tapped by the subsequent administrations to address emerging tech policy concerns.

For example, the Obama administration considerably expanded the use of multistakeholder mechanisms and other soft law tools in response to the need of oversight of fast-moving technologies. The Obama administration had many different policy governance efforts underway for specific AI technologies and concerns, including workshops and multistakeholder efforts focused on the safety, security, and privacy-related issues surrounding “big data” systems, online advertising, connected cars, drones, and more.

Whereas the Obama administration was deeper in the weeds of the policy issues associated with specific AI and machine-learning applications, the Trump administration has sought to both build on those focused efforts while also stepping back to consider AI governance at the 30,000-foot level. In essence, the AI Guidance combines some of the aspirational elements found in the Clinton Framework alongside the Obama administration’s more targeted approach to consider specific policy concerns across many different sectors and technologies.

Trump’s AI Guidance adds an element of formality to this process regarding how federal agencies should address AI developments and formulate potential policy responses. It does so by counseling humility and even potential forbearance until all the facts are in. “Fostering innovation and growth through forbearing from new regulations may be appropriate,” the memorandum says. Agencies should consider new regulation only after they have reached the decision, in light of the foregoing section and other considerations, that Federal regulation is necessary.” Again, this is very much consistent with more general regulatory guidance issued by every administration since President Reagan was in office.

Flexible, Adaptive Governance is Key

The AI Guidance foreshadows the future of not only AI governance but the governance of many other emerging technologies. Hard law will continue to provide a backstop and have a role in guiding technological developments. Toward that end, efforts like the new AI Guidance are important because it represents an effort to “regulate the regulators” by placing some ground rules on how they go about applying old law to new developments.

But soft law governance is where the real action is at, both for AI and almost all emerging technologies today. The Trump AI Guidance reflects the extent to which soft law has become the dominant governance paradigm for modern tech sectors. As my colleagues Jennifer Huddleston and Trace Mitchell have noted, soft law is already effectively the law of the land for driverless cars, for example. After years of congressional wrangling over a federal autonomous vehicle regulatory framework—one that has widespread bipartisan support, no less—we still do not have a law on the books. Instead, the Department of Transportation has been cobbling together informal “rules of the road” through informal guidance documents that have been “versioned” as if they were computer software (i.e., Version 1.0, 2.0, 3.0). Version 4.0 of the DoT guidance for automated vehicles was just released this week.

That is the same approach that the National Institute of Standards and Technology (NIST) has taken with the privacy guidelines it developed. NIST’s Privacy Framework: A Tool for Improving Privacy through Enterprise Risk Management is also versioned like software. And many other federal agencies, especially the Federal Trade Commission, have tapped a wide variety of soft law tools—such as agency workshops and workshop reports that recommended privacy best practices for various technologies. Meanwhile, the National Telecommunications and Information Administration (NTIA) has used multistakeholder processes to address privacy concerns surrounding a wide range of technologies, including drones and facial recognition. NIST, FTC, and NTIA have undertaken these informal governance efforts because, despite over a decade of debate, Congress still has not advanced comprehensive federal privacy legislation. For better or worse, soft law has filled that governance gap.

Addressing Likely Objections from Left & Right

Many people of varying ideological dispositions will object to the growing role of soft law as the primary governance tool for emerging technology policy. Some conservatives will cringe at the sound of giving regulators greater leeway to address amorphous policy concerns, fearing that it will result in unconstrained exercises of unaccountable, extra-constitutional power.

Some of those concerns are valid, but they fail to account for the fact that the prospects for agency downsizing or deregulation they prefer are extremely limited. Practically speaking, the administrative state isn’t going anywhere. In some cases, agencies can actually do some real good by encouraging innovators to think about how to “bake-in” sensible best practices to preemptively address concerns about the privacy, safety, security, and fairness of various AI systems. Better those concerns be addressed in more flexible, adaptive fashion than by a heavy-handed, overly-rigid regulatory approach. Soft law offers that possibility, even if legitimate concerns remain about agency accountability and transparency.

Many to the left of center will be critical of this governance approach as well, but on very different grounds. As Associated Press reporter Matt O’Brien notes, “the vagueness of the principles announced by the White House is unlikely to satisfy AI watchdogs who have warned of a lack of accountability as computer systems are deployed to take on human roles in high-risk social settings, such as mortgage lending or job recruitment.”

These concerns actually are addressed in several of the OSTP’s ten principles, including those which stress the need for fairness and non-discrimination, information quality, public participation, disclosure and transparency, and safety and security. Yet many on the left will claim these principles merely pay lip service to these values and that what is really needed is a full-blown regulatory regime and some sort of corresponding new federal AI agency, which would preemptively determine which AI technologies would be allowed into the wild.

Already, an Algorithmic Accountability Act was introduced in Congress last year that would ask the FTC to take a more active role in policing “inaccurate, unfair, biased, or discriminatory decisions impacting consumers” that may have resulted from “automated decision systems.” Meanwhile, some academics have called for the creation of a Federal Robotics Commission or a National Algorithmic Technology Safety Administration to preemptively oversee new AI developments.

The problem with overly-precautionary regulation of that sort could potentially unduly limit AI innovation and the many benefits it entails. There may be some AI applications that pose serious and immediate risks to humanity and which require preemptive restraints on their development and use. Autonomous military and law enforcement applications are the most obvious examples. But most AI applications do not rise to that same level of regulatory concern, and other governance approaches are required to balance the use and misuse of them. This is why a more open and flexible governance approach is needed. Moreover, the old regulatory system just cannot keep up anymore, and it is ill-suited to address most policy concerns in a timely or efficient fashion.

Cristie Ford, and advocate of greater regulatory oversight for fintech, notes in her latest book that the problem with “old-style Welfare State regulation” is that it is “a clumsy, blunt instrument for achieving regulatory objectives” due to its reliance upon “one-size-fits-all mandates, prohibitions, and penalties.” Ford acknowledges what many other regulatory advocates are reluctant to admit:  public policies toward fast-paced technology sectors can no longer be governed effectively using the Analog Era’s top-down, command-and-control regulatory processes. Far too many federal agencies rely on a “build-and-freeze model” of regulation that puts rules in stone to deal with one sets of issues one day, but then either fails to eliminate them later when they become obsolete or to reform those rules to bring them in line with new social, economic, and technical realities.

If we hope to encourage continued innovation in sectors that could produce profoundly important, life-enriching technologies, America’s regulatory approach for AI and emerging technology needs to move away from “build-and-freeze” and toward “build-and-adapt.” Regulation is still needed, but the old regulatory toolkit is badly broken. For better or worse, soft law is going to fill the resulting governance gap, regardless of objections from some on the left or the right. Pragmatic policymaking is going to carry the day for emerging technology governance.

Conclusion

The Trump Administration AI Guidance represents a continuation and extension of this trend toward more flexible, adaptive governance approaches for emerging technologies. It offers a pragmatic vision that builds on the policies and paradigms of the past, while also encouraging fresh thinking about how best to balance the need for continued innovation alongside the various concerns about disruptive technological change.

There are many challenging issues that lie ahead and the new AI Guidance cannot provide bright-line answers to all the hypothetical questions that people want answered today. No one possesses a crystal ball that will allow them to forecast the technological future. Only ongoing trial-and-error experimentation and policy improvisation will allow us to find sensible solutions. A policy approach rooted in humility, flexibility, and forbearance will help ensure that America’s regulatory policies continue to promote both innovation and the public good.

[Remarks p repared for Fifth Annual Conference on Governance of Emerging Technologies: Law, Policy & Ethics at Arizona State University, Phoenix, AZ, May 18, 2017.]

_________________

What are we to make of this peculiar new term “permissionless innovation,” which has gained increasing currency in modern technology policy discussions? And how much relevance has this notion had—or should it have—on those conversations about the governance of emerging technologies? That’s what I’d like to discuss here today.

Uncertain Origins, Unclear Definitions

I should begin by noting that while I have written a book with the term in the title, I take no credit for coining the phrase “permissionless innovation,” nor have I been able to determine who the first person was to use the term. The phrase is sometimes attributed to Grace M. Hopper, a computer scientist who was a rear admiral in the United States Navy. She once famously noted that, “It’s easier to ask forgiveness than it is to get permission.”

“Hopper’s Law,” as it has come to be known in engineering circles, is probably the most concise articulation of the general notion of “permissionless innovation” that I’ve ever heard, but Hopper does not appear to have ever used the actual phrase anywhere. Moreover, Hopper was not necessarily applying this notion to the realm of technological governance, but was seemingly speaking more generically about the benefit of trying new things without asking for the blessing of any number of unnamed authorities or overseers—which could include businesses, bosses, teachers, or perhaps even government officials.

Today, however, we most often hear the “permissionless innovation” used in discussions about the governance of information technologies as well as a wide variety of emerging technologies. Unfortunately, scholars and advocates who have suggested that permissionless innovation should serve as the governing lodestar in these areas do not always precisely define what they mean by the term.

None of them seem to be suggesting, however, that permissionless innovation is synonymous with anarchy. To the contrary, many of them are quick to note that governments will continue to have a role to play. It is even rare to see advocates of permissionless innovation in these varied contexts calling for the abolition of any laws, programs, or agencies.

Instead, it seems to be the case that most of those defenders of permissionless innovation are using the term as a sort of shorthand when what they really mean to say is something like: “give innovators a bit more breathing room,” or, “don’t rush to regulate.”

This is consistent with my own articulation of the term, which goes as follows:

“Permissionless innovation refers to the notion that experimentation with new technologies and business models should generally be permitted by default. Unless a compelling case can be made that a new invention will bring serious harm to society, innovation should be allowed to continue unabated and problems, if any develop, can be addressed later.”

Default Policy Positions

Framing the term in this fashion makes it clear that, as it pertains to technological governance, permissionless innovation is about setting our public policy defaults closer to green lights rather than red ones.

It switches the burden of proof to the opponents of ongoing technological change by asserting five things:

• First, technological innovation is the single most important determinant of long-term human well-being.
• Second, there is real value to learning through continued trial-and-error experimentation, resiliency, and ongoing adaptation to technological change.
• Third, constraints on new innovation should be the last resort, not the first. Innovation should be innocent until proven guilty.
• Fourth, as regulatory interventions are considered, policy should be based on evidence of concrete potential harm and not fear of worst-case hypotheticals.
• Fifth, and finally, where policy interventions are deemed needed, flexible, bottom-up solutions of an ex post (responsive) nature are almost always preferable to rigid, top-down controls of an ex ante (anticipatory) nature.

Shared Shortcomings of Both Visions

At least on the surface, that sort of governance vision stands in stark contrast to the “precautionary principle.” Defenders of the precautionary principle as the general default position in technology policy debates generally believe that new innovations should be curtailed or disallowed until their developers can prove that they will not cause any harm to individuals, groups, specific entities, cultural norms, or various existing laws, norms, or traditions.

That being said, I’d like to point out some of the shared shortcomings of both of these governance visions.

First, as with attempts to define the parameters of “permissionless innovation,” the precautionary principle is not always as rigid as its critics sometimes suggest. There are as many flavors of the precautionary principle as there are ice cream. Indeed, this is why many have criticized the precautionary principle not for what it says but rather for what it doesn’t say. It doesn’t tell us exactly how and when to apply precautionary measures, or how to evaluate the trade-offs associated with precaution.

This points the second and deeper underlying problem faced by advocates of both precautionary measures and permissionless innovation: Our collective inability to craft a widely-shared definition of what constitutes “technological harm” in various contexts. This is certainly not to suggest that no attempt has been made to do so. Rather, simply that we don’t seem to be any closer to concrete agreement about how or where to draw those lines.

Of course, let’s not kid ourselves into thinking that we can find bright-line answers to all these questions. After all, for many of these technological governance issues we are operating in the realm of “Level 3” or “Earth-level” systems, as Professors Allenby and Sarewitz refer to it in their book, The Techno-Human Condition. These are systems in which we deal with, as they say, “a context that is always shifting, and on meanings that are never fixed.”

That makes it even more challenging to define what we mean by “responsible innovation” or “socially desirable innovation” for purposes of determining optimal technology policy.

Risk Analysis through the Lens of Permissionless Innovation

For me, there are no easy ways out of this mess. But I do know two things for certain.

First, we must continue to refine and improve our risk analysis tools and techniques to make better determinations of when proposed interventions are sensible and cost-effective relative to the many trade-offs at work.

Again, I recognize the challenge of doing this when many of the issues and values in play are amorphous and metaphysical conflicts exist about how to even define some of these things. Most of the emerging technology policy issues I write about today, for example, involve some sort of privacy, safety, or security concern. In each case, however, very little consensus exists about what those terms even mean in varied contexts.

Nonetheless, the fact that benefit-cost analysis is hard should not serve as an excuse for failing to go through the exercise of attempting some sort of valuation of the many variables in play.

Soft Law Alternatives

The second thing I know for certain is that, due the combination of both definitional complexity regarding what constitutes technological harm, as well as the ever-accelerating pace of the so-called “pacing problem,” all roads lead back to soft law solutions instead of hard law remedies.

Last year, I had the pleasure of reading and reviewing Wendell Wallach’s new book and then having a nice conversation with him about it at Microsoft’s DC headquarters. The most interesting thing about our exchange was that, although we do not begin in the same place philosophically-speaking, we largely end up in the same place practically-speaking.

That is, there seemed to be some grudging acceptance on both our parts that “soft law” systems, multistakeholder processes, and various other informal governance mechanisms will need to fill the governance gap left by the gradual erosion of hard law.

Many other scholars, including many of you in this room, have discussed the growth of soft law mechanisms in specific contexts, but I believe we have probably failed to acknowledge the extent to which these informal governance models have already become the dominant form of technological governance, at least in the United States.

I’m currently co-authoring a very long study which documents how the Obama Administration came to rely quite heavily on multistakeholder processes, negotiated “best practices,” and industry codes of conduct as the primary governance mechanisms for a long list of emerging tech issues, including: driverless cars, commercial drones, big data, facial recognition, the Internet of Things and wearable technology, mobile medical applications, 3D printing, artificial intelligence, the Sharing Economy, and much more.

Most of these soft law processes were driven by the NTIA and FTC, but plenty of other agencies with an “N” or an “F” at the beginning of their name have undertaken some sort of soft law process, including NHTSA, the FDA, the FAA, and so on.

Now, I’m willing to bet that many of those involved in these processes who generally favor more anticipatory regulatory approaches would have preferred to start with hard law solutions to some of these issues. And I am equally certain that many of the innovators involved in those multistakeholder processes would have probably preferred not to have had to come to the table at all.

But at the end of the day, for the most part, all sides did come to the table and worked together in a good faith effort to find some rough consensus about what sort of informal guidelines would govern the future of innovation in these sectors.

The Worst of All Systems, Except All the Others

Plenty of questions remain about such soft law systems, and the irony is that defenders of both permissionless innovation and the precautionary principle will quite often be raising very similar concerns regarding the transparency, accountability, and enforceability of these systems.

But I’m inclined to believe that no matter where you sit on the permissionless vs. precautionary spectrum, and no matter what your reservations may be about it the new world of soft law governance that we find ourselves moving into, this is the future and the future is now.

Much as Churchill said of democracy being “the worst form of Government except for all those other forms that have been tried from time to time,” I think we are well on our way to a world in which soft law is the worst form of technological governance except for all those others that have been tried before.

Of course, the devil is always in the details and I suspect that we’ll have plenty of discuss and debate in that regard. Let’s get that conversation going.

Wallach’s latest book is entitled, A Dangerous Master: How to Keep Technology from Slipping beyond Our Control. And, as I’ve noted here recently, the greatly expanded second edition of my latest book, Permissionless Innovation: The Continuing Case for Comprehensive Technological Freedom, has just been released.

Of all the books of technological criticism or skepticism that I’ve read in recent years—and I have read stacks of them!— A Dangerous Master is by far the most thoughtful and interesting. I have grown accustomed to major works of technological criticism being caustic, angry affairs. Most of them are just dripping with dystopian dread and a sense of utter exasperation and outright disgust at the pace of modern technological change.

Although he is certainly concerned about a wide variety of modern technologies—drones, robotics, nanotech, and more—Wallach isn’t a purveyor of the politics of panic. There are some moments in the book when he resorts to some hyperbolic rhetoric, such as when he frets about an impending “techstorm” and the potential, as the book’s title suggests, for technology to become a “dangerous master” of humanity. For the most part, however, his approach is deeper and more dispassionate than what is found in the leading tracts of other modern techno-critics.

Many Questions, Few Clear Answers

Wallach does a particularly good job framing the major questions about emerging technologies and their effect on society. “Navigating the future of technological possibilities is a hazardous venture,” he observes. “It begins with learning to ask the right questions—questions that reveal the pitfalls of inaction, and more importantly, the passageways available for plotting a course to a safe harbor.” (p. 7) Wallach then embarks on a 260+ page inquiry that bombards the reader with an astonishing litany of questions about the wisdom of various forms of technological innovation—both large and small. While I wasn’t about to start an exact count, I would say that the number of questions Wallach poses in the book runs well into the hundreds. In fact, many paragraphs of the book are nothing but an endless string of questions.

Thus, if there is a primary weakness with A Dangerous Master, it’s that Wallach spends so much time formulating such a long list of smart and nuanced questions that some readers may come away disappointed when they do not find equally satisfying answers. On the other hand, the lack of clear answers is also completely understandable because, as Wallach notes, there really are no simple answers to most of these questions.

Just Slow Down!

Moving on to substance, let me make clear where Wallach and I generally see eye-to-eye and where we part ways.

Generally speaking, we agree about the need to come up with better “soft governance” systems for emerging technologies, which might include multistakeholder process, developer codes of conduct, sectoral self-regulation, sensible liability rules, and so on. (More on those strategies in a moment.)

But while we both believe it is wise to consider how we might “bake-in” better ethics and norms into the process of technological development, Wallach seems much more inclined than me to expect that we will be able to pre-ordain (or potentially require?) all this happens before much of this experimentation and innovation actually moves forward. Wallach opens by asking:

Determining when to bow to the judgment of experts and whether to intervene in the deployment of a new technology is certainly not easy. How can government leaders or informed citizens effectively discern which fields of research are truly promising and which pose serious risks? Do we have the intelligence and means to mitigate the serious risks that can be anticipated? How should we prepare for unanticipated risks? (p. 6)

Again, many good questions here! But this really gets to the primary difference between Wallach’s preferred approach and my own: I tend to believe that many of these things can only be worked out through ongoing trial and error, the constant reformulation of the various norms that govern the process of innovation, and the development of sensible ex post solutions to some of the most difficult problems posed by turbulent technological change.

By contrast, Wallach’s generally attitude toward technological evolution is probably best summarized by the phrases: “Slow down!” and, “Let’s have a conversation about it first!” As he puts it in his own words: “Slowing down the accelerating adoption of technology should be done as a responsible means to ensure basic human safety and to support broadly shared values.” (p. 13)

But I tend to believe that it’s not always possible to preemptively determine which innovations to slow down, or even how to determine what those “shared values” are that will help us make this determination. More importantly, I worry that there are very serious potential risks and unintended consequences associated with slowing down many forms of technological innovation, which could improve human welfare in important ways. There can be no prosperity, after all, without a certain degree of risk-taking and disruption.

Getting Out Ahead of the Pacing Problem

Yet, what concerns Wallach most is the much-discussed issue from the field of the philosophy of technology, the so-called “pacing problem.” Wallach concisely defines the pacing problem as “the gap between the introduction of a new technology and the establishment of laws, regulations, and oversight mechanisms for shaping its safe development.” (p. 251) “There has always been a pacing problem,” he notes, but he is concerned that technological innovation—especially highly disruptive and potentially uncontrollable forms of innovation—is now accelerating at an absolutely unprecedented pace.

(Just as an aside for all the philosophy nerds out there…  Such a rigid belief in the “pacing problem” represents a techno-deterministic viewpoint that is, ironically, sometimes shared by technological skeptics like Wallach as well as technological optimists like Larry Downes and even many in the middle of this debate, like Vivek Wadhwa. See, for example, The Laws of Disruption by Downes and “Laws and Ethics Can’t Keep Pace with Technology” by Wadhwa. Although these scholars approach technology ethics and politics quite differently, they all seem to believe that the pace of modern technological change is so relentless as to almost be an unstoppable force of nature. I guess the moral of the story is that, to some extent, we’re all technological determinists now!)

Despite his repeated assertions that modern technologies are accelerating at such a potentially uncontrollable pace, Wallach nonetheless hopes we can achieve some semblance of control over emerging technologies before they reach a critical “inflection point.” In the study of history and science, an inflection point generally represents a moment when a situation and trend suddenly changes in a significant way and things begin moving rapidly in a new direction. These inflections points can sometimes develop quite abruptly, ushering in major changes by creating new social, economic, or political paradigms. As it relates to technology in particular, inflection points can refer to the moment with a particular technology achieves critical mass in terms of adoption or, more generally, to the time when that technology begins to profoundly transform the way individuals and institutions act.

Another related concept that Wallach discusses is the so-called “Collingridge dilemma,” which refers to the notion that it is difficult to put the genie back in the bottle once a given technology has reached a critical mass of public adoption or acceptance. The concept is named after David Collingridge, who wrote about this in his 1980 book, The Social Control of Technology. “The social consequences of a technology cannot be predicated early in the life of the technology,” Collingridge argued. “By the time undesirable consequences are discovered, however, the technology is often so much part of the whole economics and social fabric that its control is extremely difficult.”

On “Having a Discussion” & Coming Up with “a Broad Plan”

These related concepts of inflection points and the Collingridge dilemma constitute the operational baseline of Wallach’s worldview. “In weighing speedy development against long-term risks, speedy development wins,” he worries. “This is particularly true when the risks are uncertain and the perceived benefits great.” (p. 85)

Consequently, throughout his book, Wallach pleads with us to take what I will call Technological Time Outs. He says we need to pause at times so that we can have “a full public discussion” (p. 13) and make sure there is a “broad plan in place to manage our deployment of new technologies” (p. 19) to make sure that innovation happens only at “a humanly manageable pace” (p. 261) “to fortify the safety of people affected by unpredictable disruptions.” (p. 262) Wallach’s call for Technological Time Outs is rooted in his belief that “the accelerating pace [of modern technological innovation] undermines the quality of each of our lives.” (p. 263)

That is Wallach’s weakest assertion in the book and he doesn’t really offer much evidence to prove that the velocity of modern technological is hurting us rather than helping us, as many of us believe. Rather, he treats it as a widely accepted truism that necessitates some sort of collective effort to slow things down if the proverbial genie is about to exit the bottle, or to make sure those genies don’t get out of their bottles without a lot of preemptive planning regarding how they are to be released into the world. In the following passage on pg. 72, Wallach very succinctly summarizes his approach recommended throughout A Dangerous Master:

this book will champion the need for more upstream governance: more control over the way that potentially harmful technologies are developed or introduced into the larger society. Upstream management is certainly better than introducing regulations downstream, after a technology is deeply entrenched or something major has already gone wrong. Yet, even when we can access risks, there remain difficulties in recognizing when or determining how much control should be introduced. When does being precautionary make sense, and when is precaution an over-reaction to the risks? (p. 72)

Those who have read my Permissionless Innovation book will recall that I open by framing innovation policy debates in almost exactly the same way as Wallach suggests in that last line above. I argue in the first lines of my book that:

The central fault line in innovation policy debates today can be thought of as ‘the permission question.’  The permission question asks: Must the creators of new technologies seek the blessing of public officials before they develop and deploy their innovations? How that question is answered depends on the disposition one adopts toward new inventions and risk-taking, more generally.  Two conflicting attitudes are evident. One disposition is known as the ‘precautionary principle.’ Generally speaking, it refers to the belief that new innovations should be curtailed or disallowed until their developers can prove that they will not cause any harm to individuals, groups, specific entities, cultural norms, or various existing laws, norms, or traditions. The other vision can be labeled ‘permissionless innovation.’ It refers to the notion that experimentation with new technologies and business models should generally be permitted by default. Unless a compelling case can be made that a new invention will bring serious harm to society, innovation should be allowed to continue unabated and problems, if any develop, can be addressed later.

So, by contrasting these passages, you can see what I am setting up here is a clash of visions between what appears to be Wallach’s precautionary principle-based approach versus my own permissionless innovation-focused worldview.

How Much Formal Precaution?

But that would be a tad bit too simplistic because just a few paragraphs after Wallach makes the statement just above about “upstream management” being superior to ex post solutions formulated “after a technology is deeply entrenched,” Wallach begins slowly backing away from an overly-rigid approach to precautionary principle-based governance of technological processes and systems.

He admits, for example, that “precautionary measures in the form of regulations and governmental oversight can slow the development of research whose overall society impact will be beneficial,” (p. 26) and that can “be costly” and “slow innovation.” For countries, Wallach admits, this can have real consequences because “Countries with more stringent precautionary policies are at a competitive disadvantage to being the first to introduce a new tool or process.” (p. 74)

So, he’s willing to admit that what we might call a hard precautionary principle usually won’t be sensible or effective in practice, but he is far more open to soft precaution. But this is where real problems begin to develop with Wallach’s approach, and it presents us with a chance to turn the tables on him a bit and begin posing some serious questions about his vision for governing technology.

Much of what follows below are my miscellaneous ramblings about the current state of the intellectual dialogue about tech ethics and technological control efforts. I have discussed these issues at greater length in my new book as well as a series of essays here in past years, most notably: “On the Line between Technology Ethics vs. Technology Policy; “What Does It Mean to “Have a Conversation” about a New Technology?”; and, “Making Sure the “Trolley Problem” Doesn’t Derail Life-Saving Innovation.”

As I’ve argued in those and other essays, my biggest problem with modern technological criticism is that specifics are in scandalously short supply in this field! Indeed, I often find the lack of details in this arena to be utterly exasperating. Most modern technological criticism follows a simple formula:

TECHNOLOGY –>> POTENTIAL PROBLEMS –>> DO SOMETHING!

But almost all the details come in the discussion about the nature of the technology in question and the apparent many problems associated with it. Far, far less thought goes into the “DO SOMETHING!” part of the critics’ work. One reason for that is probably self-evident: There are no easy solutions. Wallach admits as much at many junctures throughout the book. But that doesn’t excuse the need for the critics to give us a more concrete blueprint for identifying and then potentially rectifying the supposed problems.

Of course, the other reason that many critics are short of specifics is because what they really mean when they quip how much we need to “have a conversation” about a new disruptive technology is that we need to have a conversation about stopping that technology.

Where Shall We Draw the Line between Hard and Soft Law?

But this is what I found most peculiar about Wallach’s book: He never really gives us a good standard by which to determine when we should look to hard governance (traditional top-down regulation) versus soft governance (more informal, bottom-up and non-regulatory approaches).

On one hand, he very much wants society to exercise greatly restraint and precaution when it comes to many of the technologies he and others worry about today. Again, he’s particularly concerned about the potential runaway development and use of drones, genetic editing, nanotech, robotics, and artificial intelligence. For at least one class of robotics—autonomous military robots—Wallach does call for immediate policy action in the form of an Executive Order to ban “killer” autonomous systems. (Incidentally, there’s also a major effort underway called the “Campaign to Stop Killer Robots” that aims to make such a ban part of international law through a multinational treaty.)

But Wallach also acknowledges the many trade-offs associated with efforts to preemptively controls on robotics and other technology. Perhaps for that reason, Wallach doesn’t develop a clear test for when the Precautionary Principle should be applied to new forms of innovation.

Clearly there are times when it is appropriate, although I believe it is only in an extremely narrow subset of cases. In the 2 ^nd Edition of my Permissionless Innovation book, I tried to offer a rough framework for when formal precautionary regulation (i.e., highly-restrictive policy defaults are necessary, such as operational restrictions, licensing requirements, research limitations, or even formal bans) might be necessary. I do not want to interrupt the flow of this review of Wallach’s book too much, so I have decided to just cut-and-paste that portion of Chapter 3 of my book (“When Does Precaution Make Sense?”) down below as an appendix to this essay.

The key takeaway of that passage from my book is that all of us who study innovation policy and the philosophy of technology—Wallach, myself, the whole darn movement—have done a remarkably poor job being specific about precisely when formal policy precaution is warranted. What is the test? All too often, we get lazy and apply what we might call an “I-Know-It-When-I-See-It” standard. Consider the possession of bazookas, tanks, and uranium. Almost all of us would agree that citizens should not be allowed to possess or use such things. Why? Well, it seems obvious, right? They just shouldn’t! But what is the exact standard we use to make that determination.

In coming years, I plan on spending a lot more time articulating a better test by which Precautionary Principle-based policies could be reasonably applied. Those who know me may be taken aback by what I just said. After all, I’ve spend many years explaining why Precautionary Principle-based thinking threatens human prosperity and should be rejected in the vast majority of cases. But that doesn’t excuse the lack of a serious and detailed exploration of the exact standard by which we determine when we should impose some limits on technological innovation.

Generally speaking, while I strongly believe that “permissionless innovation” should remain the policy default for most technologies, there certainly exists some scenarios where the threat of harm associated with a new innovation might be highly probable, tangible, immediate, irreversible, and catastrophic in nature. If so, that could qualify it for at least a light version of the Precautionary Principle. In a future paper or book chapter I’m just now starting to research, I hope to fuller develop those qualifiers and formulate a more robust test around them.

I would have very much liked to see Wallach articulate and defend a test of his own for when formal precaution would make sense. And, by extension, when should we default to soft precaution, or soft law and informal governance mechanisms for emerging technologies.

We turn to that issue next.

Toward Soft Governance & the Engineering of Better Technological Ethics

Even though Wallach doesn’t provide us with a test for determining when precaution makes sense or when we should instead default to soft governance, he does a much better job explaining the various models of soft law or informal governance that might help us deal with the potential negative ramifications of highly disruptive forms of technological change.

What Wallach proposes, in essence, is that we bake a dose of precautionary directly into the innovation process through a wide variety of informal governance/oversight mechanisms. “By embedding shared values in the very design of new tools and techniques, engineers improve the prospect of a positive outcome,” he claims. “The upstream embedding of shared values during the design process can ease the need for major course adjustments when it’s often too late.” (p. 261)

Wallach’s favored instrument of soft governance is what he refers to as “Governance Coordinating Committees” (GCCs). These Committees would coordinate “the separate initiatives by the various government agencies, advocacy groups, and representatives of industry” who would serve as “issue managers for the comprehensive oversight of each field of research.” (p. 250) He elaborates and details the function of GCCs as follows:

These committees, led by accomplished elders who have already achieved wide respect, are meant to work together with all the interested stakeholders to monitor technological development and formulate solutions to perceived problems. Rather than overlap with or function as a regulatory body, the committee would work together with existing institutions. (p. 250-51)

Wallach discussed the GCC idea in much greater detail in a 2013 book chapter he penned with Gary E. Marchant for a collected volume of essays on Innovative Governance Models for Emerging Technologies. (I highly recommend you pick up that book if you can afford it! Many terrific essays in that book on these issues.) In their chapter, Marchant and Wallach specify some of the soft law mechanisms we might use to instill a bit of precaution preemptively. These mechanisms include: “codes of conduct, statements of principles, partnership programs, voluntary programs and standards, certification programs and private industry initiatives.”

If done properly, GCCs could provide exactly the sort of wise counsel and smart recommendations that Wallach desires. In my book and many law review articles on various disruptive technologies, I have endorsed many of the ideas and strategies Wallach identifies. I’ve also stressed the importance of many other mechanisms, such as education and empowerment-based strategies that could help the public learn to cope with new innovations or use them appropriately. In addition, I’ve highlighted the many flexible, adaptive ex post remedies that can help when things go wrong. Those mechanisms include common law remedies such as product defects law, various torts, contract law, property law, and even class action lawsuits. Finally, I have written extensively about the very active role played by the Federal Trade Commission (FTC) and other consumer protection agencies, which have broad discretion to police “unfair and deceptive practices” by innovators.

Moreover, we already have a quasi-GCC model developing today with the so-called “multistakeholder governance” model that is often used in both informal and formal ways to handle many emerging technology policy issues.  The Department of Commerce (the National Telecommunications and Information Administration in particular) and the FTC have already developed many industry codes of conduct and best practices for technologies such as biometrics, big data, the Internet of Things, online advertising, and much more. Those agencies and others (such as the FDA and FAA) are continuing to investigate other codes or guidelines for things like advanced medical devices and drones, respectively. Meanwhile, I’ve heard other policymakers and academics float the idea of “digital ombudsmen,” “data ethicists,” and “private IRBs” (institutional review boards) as other potential soft law solutions that technology companies might consider. Perhaps going forward, many tech firms will have Chief Ethical Officers just as many of them today have Chief Privacy Officers or Chief Security Officers.

In other words, there’s already a lot of “soft law” activities going on in this space. And I haven’t even begun an inventory of the many other bodies or groups that already exist in each sector today that has set forth their own industry self-regulatory codes, but they exist in almost every field that Wallach worries about.

So, I’m not sure how much his GCC idea will add to this existing mix, but I would not be opposed to them playing the sort of coordinating “issue manager” role he describes. But I still have many questions about GCC’s, including:

• How many of them are needed and how we will know which one is the definitive GCC for each sector or technology?
• If they are overly formal in character and dominated by the most vociferous opponents of any particular technology, a real danger exists that a GCC could end up granting a small cabal a “heckler’s veto” over particular forms of innovation.
• Alternatively, the possibility of “regulatory capture” could be a problem for some GCCs if incumbent companies come to dominate their membership.
• Even if everything went fairly smoothly and the GCCs produced balanced reports and recommendations, future developers might wonder if and why they are to be bound by older guidelines.
• And if those future developers choose not to play by the same set of guidelines, what’s the penalty for non-compliance?
• And how are such guidelines enforced in a world where what I’ve called “global innovation arbitrage” is an increasing reality?

Challenging Questions for Both Hard and Soft Law

To summarize, whether we are speaking of “hard” or “soft” law approaches to technological governance, I am just not nearly as optimistic as Wallach seems to be that we will be able to find consensus on these three things:

(1) what constitutes “harm” in many of these circumstances;

(2) which “shared values” should prevail when “society” debates the shaping of ethics or guiding norms for emerging technologies but has highly contradictory opinions about those values (consider online privacy as a good example, where many people enjoy hyper-sharing while other demand hyper-privacy); and,

(3) that we can create a legitimate “governing body” (or bodies) that will be responsible for formulating these guidelines in a fair way without completely derailing the benefits of innovation in new fields and also remaining relevant for very long.

Nonetheless, as he and others have suggested, the benefit of adopting a soft law/informal governance approach to these issues is that it at least seeks to address these questions in more flexible and adaptive fashion. As I noted in my book, traditional regulatory systems “tend to be overly rigid, bureaucratic, inflexible, and slow to adapt to new realities. They focus on preemptive remedies that aim to predict the future, and future hypothetical problems that may not ever come about. Worse yet, administrative regulation generally preempts or prohibits the beneficial experiments that yield new and better ways of doing things.” ( Permissionless Innovation, p. 120)

So, despite the questions I have raised here, I welcome the more flexible soft law approach that Wallach sets forth in his book. I think it represents a far more constructive way forward when compared to the opposite “top-down” or “command-and-control” regulatory systems of the past. But I very much want to make sure that even these new and more flexible soft law approaches leave plenty of breathing room for ongoing trial-and-error experimentation with new technologies and systems.

Conclusion

In closing, I want to reiterate that not only did I appreciate the excellent questions raised by Wendell Wallach in A Dangerous Master, but I take them very seriously. When I sat down to revise and expand my Permissionless Innovation book last year, I decided to include this warning from Wallach in my revised preface: “The promoters of new technologies need to speak directly to the disquiet over the trajectory of emerging fields of research. They should not ignore, avoid, or superficially dampen criticism to protect scientific research.” (p. 28–9)

As I noted, in response to Wallach: “I take this charge seriously, as should others who herald the benefits of permissionless innovation as the optimal default for technology policy. We must be willing to take on the hard questions raised by critics and then also offer constructive strategies for dealing with a world of turbulent technological change.”

Serious questions deserve serious answers. Of course, sometimes those posing those questions fail to provide many answers of their own! Perhaps it is because they believe the questions answer themselves. Other times, it’s because they are willing to admit that easy answers to these questions typically prove quite elusive. In Wallach’s case, I believe it’s more the latter.

To wrap up, I’ll just reiterated that both Wallach and I share a common desire to find solutions to the hard questions about technological innovation. But the crucial question that probably separates his worldview and my own is this: Whether we are talking about hard or soft governance, how much faith should we place in preemptive planning vs. ongoing trial and error experimentation to solve technological challenges? Wallach is more inclined to believe we can divine these things with the sagacious foresight of “accomplished elders” and technocratic “issue managers,” who will help us slow things down until we figure out how to properly ease a new technology into society (if at all). But I believe that the only way we will find many of the answers we are searching for is by allowing still more experimentation with the very technologies that he and others seek to control the development of. We humans are outstanding problem-solvers and have the uncanny ability among all mammals to adapt to changing circumstances. We roll with the punches, learn from them, and become more resilient in the process. As I noted in my 2014 essay, “Muddling Through: How We Learn to Cope with Technological Change”:

we modern pragmatic optimists must continuously point to the unappreciated but unambiguous benefits of technological innovation and dynamic change. But we should also continue to remind the skeptics of the amazing adaptability of the human species in the face of adversity. [. . .] Humans have consistently responded to technological change in creative, and sometimes completely unexpected ways. There’s no reason to think we can’t get through modern technological disruptions using similar coping and adaptation strategies.

Will the technologies that Wallach fears bring about a “techstorm” that overwhelms our culture, our economy, and even our very humanity? It’s certainly possible, and we should continue to seriously discuss the issues that he and other skeptics raise about our expanding technological capabilities and the potential for many of them to do great harm. Because some of them truly could.

But it is equally plausible—in fact, some of us would say, highly probable—that instead of overwhelming us, we learn how to bend these new technological capabilities to our will and make them work for our collective benefit. Instead of technology becoming “a dangerous master,” we will instead make it our helpful servant, just as we have so many times before.

APPENDIX: When Does Precaution Make Sense?

[excerpt from chapter 3 of Permissionless Innovation: The Continuing Case for Comprehensive Technological Freedom. Footnotes omitted. See book for all references.]

But aren’t there times when a certain degree of precautionary policymaking makes good sense? Indeed, there are, and it is important to not dismiss every argument in favor of precautionary principle–based policymaking, even though it should not be the default policy rule in debates over technological innovation.

The challenge of determining when precautionary policies make sense comes down to weighing the (often limited) evidence about any given technology and its impact and then deciding whether the potential downsides of unrestricted use are so potentially catastrophic that trial-and-error experimentation simply cannot be allowed to continue. There certainly are some circumstances when such a precautionary rule might make sense. Governments restrict the possession of uranium and bazookas, to name just two obvious examples.

Generally speaking, permissionless innovation should remain the norm in the vast majority of cases, but there will be some scenarios where the threat of tangible, immediate, irreversible, catastrophic harm associated with new innovations could require at least a light version of the precautionary principle to be applied.  In these cases, we might be better suited to think about when an “anti-catastrophe principle” is needed, which narrows the scope of the precautionary principle and focuses it more appropriately on the most unambiguously worst-case scenarios that meet those criteria.

But most cases don’t fall into this category. Instead, we generally allow innovators and consumers to freely experiment with technologies, and even engage in risky behaviors, unless a compelling case can be made that precautionary regulation is absolutely necessary.  How is the determination made regarding when precaution makes sense? This is where the role of benefit-cost analysis (BCA) and regulatory impact analysis is essential to getting policy right.  BCA represents an effort to formally identify the tradeoffs associated with regulatory proposals and, to the maximum extent feasible, quantify those benefits and costs.  BCA generally cautions against preemptive, precautionary regulation unless all other options have been exhausted—thus allowing trial-and-error experimentation and “learning by doing” to continue. (The mechanics of BCA are discussed in more detail in section VII.)

This is not the end of the evaluation, however. Policymakers also need to consider the complexities associated with traditional regulatory remedies in a world where technological control is increasingly challenging and quite costly. It is not feasible to throw unlimited resources at every problem, because society’s resources are finite.  We must balance risk probabilities and carefully weigh the likelihood that any given intervention has a chance of creating positive change in a cost-effective fashion.  And it is also essential to take into account the potential unintended consequences and long-term costs of any given solution because, as Harvard law professor Cass Sunstein notes, “it makes no sense to take steps to avert catastrophe if those very steps would create catastrophic risks of their own.”  “The precautionary principle rests upon an illusion that actions have no consequences beyond their intended ends,” observes Frank B. Cross of the University of Texas. But “there is no such thing as a risk-free lunch. Efforts to eliminate any given risk will create some new risks,” he says.

Oftentimes, after working through all these considerations about whether to regulate new technologies or technological processes, the best solution will be to do nothing because, as noted throughout this book, we should never underestimate the amazing ingenuity and resiliency of humans to find creative solutions to the problems posed by technological change.  (Section V discusses the importance of individual and social adaptation and resiliency in greater detail.) Other times we might find that, while some solutions are needed to address the potential risks associated with new technologies, nonregulatory alternatives are also available and should be given a chance before top-down precautionary regulations are imposed. (Section VII considers those alternative solutions in more detail.)

Finally, it is again essential to reiterate that we are talking here about the dangers of precautionary thinking as a public policy prerogative—that is, precautionary regulations that are mandated and enforced by government officials. By contrast, precautionary steps may be far more wise when undertaken in a more decentralized manner by individuals, families, businesses, groups, and other organizations. In other words, as I have noted elsewhere in much longer articles on the topic, “there is a different choice architecture at work when risk is managed in a localized manner as opposed to a society-wide fashion,” and risk-mitigation strategies that might make a great deal of sense for individuals, households, or organizations, might not be nearly as effective if imposed on the entire population as a legal or regulatory directive.

Finally, at times, more morally significant issues may exist that demand an even more exhaustive exploration of the impact of technological change on humanity. Perhaps the most notable examples arise in the field of advance medical treatments and biotechnology. Genetic experimentation and human cloning, for example, raise profound questions about altering human nature or abilities as well as the relationship between generations.

The case for policy prudence in these matters is easier to make because we are quite literally talking about the future of what it means to be human.  Controversies have raged for decades over the question of when life begins and how it should end. But these debates will be greatly magnified and extended in coming years to include equally thorny philosophical questions.  Should parents be allowed to use advanced genetic technologies to select the specific attributes they desire in their children? Or should parents at least be able to take advantage of genetic screening and genome modification technologies that ensure their children won’t suffer from specific diseases or ailments once born?

Outside the realm of technologically enhanced procreation, profound questions are already being raised about the sort of technological enhancements adults might make to their own bodies. How much of the human body can be replaced with robotic or bionic technologies before we cease to be human and become cyborgs?  As another example, “biohacking”—efforts by average citizens working together to enhance various human capabilities, typically by experimenting on their own bodies —could become more prevalent in coming years.  Collaborative forums, such as Biohack.Me, already exist where individuals can share information and collaborate on various projects of this sort.  Advocates of such amateur biohacking sometimes refer to themselves as “grinders,” which Ben Popper of the Verge defines as “homebrew biohackers [who are] obsessed with the idea of human enhancement [and] who are looking for new ways to put machines into their bodies.”

These technologies and capabilities will raise thorny ethical and legal issues as they advance. Ethically, they will raise questions of what it means to be human and the limits of what people should be allowed to do to their own bodies. In the field of law, they will challenge existing health and safety regulations imposed by the FDA and other government bodies.

Again, most innovation policy debates—including most of the technologies discussed throughout this book—do not involve such morally weighty questions. In the abstract, of course, philosophers might argue that every debate about technological innovation has an impact on the future of humanity and “what it means to be human.” But few have much of a direct influence on that question, and even fewer involve the sort of potentially immediate, irreversible, or catastrophic outcomes that should concern policymakers.

In most cases, therefore, we should let trial-and-error experimentation continue because “experimentation is part and parcel of innovation” and the key to social learning and economic prosperity.  If we froze all forms of technological innovation in place while we sorted through every possible outcome, no progress would ever occur. “Experimentation matters,” notes Harvard Business School professor Stefan H. Thomke, “because it fuels the discovery and creation of knowledge and thereby leads to the development and improvement of products, processes, systems, and organizations.”

Of course, ongoing experimentation with new technologies always entails certain risks and potential downsides, but the central argument of this book is that (a) the upsides of technological innovation almost always outweigh those downsides and that (b) humans have proven remarkably resilient in the face of uncertain, ever-changing futures.

In sum, when it comes to managing or coping with the risks associated with technological change, flexibility and patience is essential. One size most certainly does not fit all. And one-size-fits-all approaches to regulating technological risk are particularly misguided when the benefits associated with technological change are so profound. Indeed, “[t]echnology is widely considered the main source of economic progress”; therefore, nothing could be more important for raising long-term living standards than creating a policy environment conducive to ongoing technological change and the freedom to innovate.

I recently finished  Learning by Doing: The Real Connection between Innovation, Wages, and Wealth , by James Bessen of the Boston University Law School. It’s a good book to check out if you are worried about whether workers will be able to weather this latest wave of technological innovation.  One of the key insights of Bessen’s book is that, as with previous periods of turbulent technological change, today’s workers and businesses will obviously need find ways to adapt to rapidly-changing marketplace realities brought on by the Information Revolution, robotics, and automated systems.

That sort of adaptation takes time, but for technological revolutions to take hold and have meaningful impact on economic growth and worker conditions, it requires that large numbers of ordinary workers acquire new knowledge and skills, Bessen notes. But, “that is a slow and difficult process, and history suggests that it often requires social changes supported by accommodating institutions and culture.” (p 223) That is not a reason to resist disruptive forms of technological change, however. To the contrary, Bessen says, it is crucial to allow ongoing trial-and-error experimentation and innovation to continue precisely because it represents a learning process which helps people (and workers in particular) adapt to changing circumstances and acquire new skills to deal with them. That, in a nutshell, is “learning by doing.” As he elaborates elsewhere in the book:

Major new technologies become ‘revolutionary’ only after a long process of learning by doing and incremental improvement. Having the breakthrough idea is not enough. But learning through experience and experimentation is expensive and slow. Experimentation involves a search for productive techniques: testing and eliminating bad techniques in order to find good ones. This means that workers and equipment typically operate for extended periods at low levels of productivity using poor techniques and are able to eliminate those poor practices only when they find something better. (p. 50)

Luckily, however, history also suggests that, time and time again, that process has happened and the standard of living for workers and average citizens alike improved at the same time.

Of course, that won’t stop some from proclaiming that,  This time it’s different! Indeed, we’re hearing increasing concerns today about the “rise of the robots,” and the general negative impact of automation on the workforce.

But these concerns are really nothing new. “There have been periodic warnings in the last two centuries that automation and new technology were going to wipe out large numbers of middle class jobs,” notes MIT economist David H. Autor. Luckily, those dire predictions have not come to pass. The reason was because short-sighted skeptics failed to appreciate how as new technologies obliterated old businesses and jobs, it simultaneously opened up many more opportunities that were impossible to predict in advance. For every factory worker that lost a job due to technological innovation, new jobs opened up in entirely new sectors that usually offered workers better wages, a safer work environment, and more leisure time. And society clearly benefited in many other ways.

In a new essay for  The Journal of Economic Perspectives on “The History of Technological Anxiety and the Future of Economic Growth: Is This Time Different?” Joel Mokyr, Chris Vickers, and Nicolas L. Ziebarth, note that “Discussions of how technology may affect labor demand are often focused on existing jobs, which can offer insights about which occupations may suffer the greatest dislocation, but offer much less insight about the emergence of as-yet-nonexistent occupations of the future.” They continue on to note that:

In the end, the fears of the Luddites that machinery would impoverish workers were not realized, and the main reason is well understood. The mechanization of the early 19th century could only replace a limited number of human activities. At the same time, technological change increased the demand for other types of labor that were complementary to the capital goods embodied in the new technologies. This increased demand for labor included such obvious jobs as mechanics to fix the new machines, but it extended to jobs for supervisors to oversee the new factory system and accountants to manage enterprises operating on an unprecedented scale. More importantly, technological progress also took the form of product innovation, and thus created entirely new sectors for the economy, a development that was essentially missed in the discussions of economists of this time.

And despite a resurgence of automation anxiety in recent years, that historic trend still generally holds true. In late 2014, economists at Deloitte LLP published a sweeping survey of the impact of technology and jobs over the past 200 years and found that “Technology has transformed productivity and living standards, and, in the process, created new employment in new sectors.” This is because human needs and wants constantly change and, therefore, “The stock of work in the economy is not fixed; the last 200 years demonstrates that when a machine replaces a human, the result, paradoxically, is faster growth and, in time, rising employment.” And they conclude that: “Machines will take on more repetitive and laborious tasks, but seem no closer to eliminating the need for human labour than at any time in the last 150 years. It is not hard to think of pressing, unmet needs even in the rich world: the care of the elderly and the frail, lifetime education and retraining, health care, physical and mental well-being.”

While it is easy for critics to highlight disruptions in some notable sectors where machines replaced human labor, fewer news reports or panicky books discuss the many new sectors where people have found new opportunities. Again, the historical evidence suggests that there are good reasons to have faith that humans will once again muddle through and prevail in the face of turbulent, disruptive change. As venture capitalist Marc Andreessen has noted when addressing the fear that automation is running amuck and that robots will eat all our jobs,

We have no idea what the fields, industries, businesses, and jobs of the future will be. We just know we will create an enormous number of them. Because if robots and AI replace people for many of the things we do today, the new fields we create will be built on the huge number of people those robots and AI systems made available. To argue that huge numbers of people will be available but we will find nothing for them (us) to do is to dramatically short human creativity. And I am way long human creativity.

Some tech critics may reject Andreessen’s bullish optimism about human resiliency, but real-world evidence already supports that his conclusion that we’ll learn to adapt to a world full of robots and robotic systems. A 2015 economic analysis from Colin Lewis, a behavioral economist who runs Robotenomics, showed that “despite the headlines, companies that have installed industrial robots are actually increasingly employing more people whilst at the same time adding more robots.” His research revealed that 1.25 million new jobs had been added by companies that make extensive use of industrial robots over the previous 6 years. He also found that this trend held among more recent disruptive firms like Amazon and Tesla Motors, but also older and more established companies like Chrysler, Daimler, Philips Electronics and others.

So, it’s worth keeping these facts in mind next time you read an article or book that declares that the sky is falling and that technological innovation is going to destroy labor markets and living standards. The entirety of human history points in the opposite direction. We should be bullish about our ability to muddle through tough times of technological change and flourish in the long run.

[Last updated July 2021.]

I spend a lot of time reading books and essays about technology; more specifically, books and essays about technology history and criticism. Yet, I am often struck by how few of the authors of these works even bother defining what they mean by “technology.” I find that frustrating because, if you are going to make an attempt to either study or critique a particular technology or technological practice or development, then you probably should take the time to tell us how broadly or narrowly you are defining the term “technology” or “technological process.”

Of course, it’s not easy. “In fact, technology is a word we use all of the time, and ordinarily it seems to work well enough as a shorthand, catch-all sort of word,” notes the always-insightful Michael Sacasas in his essay “Traditions of Technological Criticism.” “That same sometimes useful quality, however, makes it inadequate and counter-productive in situations that call for more precise terminology,” he says.

Quite right, and for a more detailed and critical discussion of how earlier scholars, historians, and intellectuals have defined or thought about the term “technology,” you’ll want to check out Michael’s other recent essay, “What Are We Talking About When We Talk About Technology?” which preceded the one cited above. We don’t always agree on things — in fact, I am quite certain that most of my comparatively amateurish work must make his blood boil at times! — but you won’t find a more thoughtful technology scholar alive today than Michael Sacasas. If you’re serious about studying technology history and criticism, you should follow his blog and check out his book, The Tourist and The Pilgrim: Essays on Life and Technology in the Digital Age, which is a collection of some of his finest essays.

Anyway, for what it’s worth, I figured I would create this post to list some of the more interesting definitions of “technology” that I have uncovered in my own research. I suspect I will add to it in coming months and years, so please feel free to suggest other additions since I would like this to be a useful resource to others.

I figure the easiest thing to do is to just list the definitions by author. There’s no particular order here, although that might change in the future since I could arrange this chronologically and push the inquiry all the way back to how the Greeks thought about the term (the root term “techne,” that is). But for now this collection is a bit random and incorporates mostly modern conceptions of “technology” since the term didn’t really gain traction until relatively recent times.

Also, I’ve not bothered critiquing any particular definition or conception of the term, although that may change in the future, too. (I did, however, go after a few modern tech critics briefly in my recent booklet, “Permissionless Innovation: The Continuing Case for Comprehensive Technological Freedom.” So, you might want to check that out for more on how I feel, as well as my old essays, “What Does It Mean to ‘Have a Conversation’ about a New Technology?” and, “On the Line between Technology Ethics vs. Technology Policy.”)

So, I’ll begin with two straight-forward definitions from the Merriam-Webster and Oxford dictionaries and then bring in the definitions from various historians and critics.

Merriam-Webster Dictionary

Technology (noun):

1)     (a): the practical application of knowledge especially in a particular area; (b): a capability given by the practical application of knowledge

2)      a manner of accomplishing a task especially using technical processes, methods, or knowledge.

3)      the specialized aspects of a particular field of endeavor.

Oxford Dictionary

Technology (noun):

1)      The application of scientific knowledge for practical purposes, especially in industry.

2)      Machinery and devices developed from scientific knowledge.

3)      The branch of knowledge dealing with engineering or applied sciences.

Emmanuel Mesthene

My personal favorite definition of the term comes from Emmanuel G. Mesthene’s terrific little 1970 book, Technological Change: Its Impact on Man and Society:

“we define technology as the organization of knowledge for the achievement of practical purposes.”

John Kenneth Galbraith

A very similar definition to Mesthene’s was employed by Galbraith in his 1967 book  The New Industrial State:

“Technology means the systematic application of scientific or other organized  knowledge to practical tasks.”

Thomas P. Hughes

I have always loved the opening passage from Thomas Hughes’s 2004 book, Human-Built World: How to Think about Technology and Culture:

“Technology is messy and complex. It is difficult to define and to understand. In its variety, it is full of contradictions, laden with human folly, saved by occasional benign deeds, and rich with unintended consequences.” (p. 1) “Defining technology in its complexity,” he continued, “is as difficult as grasping the essence of politics.” (p. 2)

So true! Nonetheless, Hughes went on to offer his own definition of technology as:

“a creativity process involving human ingenuity.” (p. 3)

Interestingly, in another book, American Genesis: A Century of Invention and Technological Enthusiasm, 1870-1970, he offered a somewhat different definition:

“Technology is the effort to organize the world for problem solving so that goods and services can be invented, developed, produced, and used.” (p. 6, 2004 ed., emphasis in original.)

W. Brian Arthur

In his 2009 book, The Nature of Technology: What It Is and How It Evolves, W. Brian Arthur sketched out three conceptions of technology.

1)      “The first and most basic one is a technology is a means to fulfill a human purpose. … As a means, a technology may be a method or process or device… Or it may be complicated… Or it may be material… Or it may be nonmaterial. Whichever it is, it is always a means to carry out a human purpose.” 2)      “The second definition is a plural one: technology as an assemblage of practices and components.” 3)      “I will also allow a third meaning. This technology as the entire collection of devices and engineering practices available to a culture.” (p. 28, emphasis in original.) 

Alfred P. Sloan Foundation / Richard Rhodes

In his 1999 book, Visions of Technology: A Century Of Vital Debate About Machines Systems And The Human World, Pulitizer Prize-winning historian Richard Rhodes assembled a wonderful collection of essays about technology that spanned the entire 20^th century. It’s a terrific volume to have on your bookshelf if want a quick overview of how over a hundred leading scholars, critics, historians, scientists, and authors thought about technology and technological advances.

The collection kicked off with a brief preface from the Alfred P. Sloan Foundation (no specific Foundation author was listed) that included one of the most succinct definitions of the term you’ll ever read:

“Technology is the application of science, engineering and industrial organization to create a human-build world.” (p. 19)

Just a few pages later, however, Rhodes notes that is probably too simplistic:

“Ask a friend today to define technology and you might hear words like ‘machines,’ ‘engineering,’ ‘science.’ Most of us aren’t even sure where science leaves off and technology begins. Neither are the experts.”

Again, so true!

Joel Mokyr

Lever of Riches: Technological Creativity and Economic Progress(1990) by Joel Mokyr is one of the most readable and enjoyable histories of technology you’ll ever come across. I highly recommend it. [My thanks to my friend William Rinehart for bringing the book to my attention.]  In Lever of Riches, Mokyr defines “technological progress” as follows:

“By technological progress I mean any change in the application of information to the production process in such a way as to increase efficiency, resulting either in the production of a given output with fewer resources (i.e., lower costs), or the production of better or new products.” (p. 6)

Edwin Mansfield

You’ll find definitions of both “technology” and “technological change” in Edwin Mansfield’s Technological Change: An Introduction to a Vital Area of Modern Economics (1968, 1971):

“Technology is society’s pool of knowledge regarding the industrial arts. It consists of knowledge used by industry regarding the principles of physical and social phenomena… knowledge regarding the application of these principles to production… and knowledge regarding the day-to-day operations of production…” “Technological change is the advance of technology, such advance often taking the form of new methods of producing existing products, new designs which enable the production of products with important new characteristics, and new techniques of organization, marketing, and management.” (p. 9-10)

Read Bain

In his December 1937 essay in Vol. 2, Issue No. 6 of the American Sociological Review, “Technology and State Government,” Read Bain said:

 “technology includes all tools, machines, utensils, weapons, instruments, housing, clothing, communicating and transporting devices and the skills by which we produce and use them.” (p. 860)

[My thanks to Jasmine McNealy for bringing this one to my attention.]

David M. Kaplan

Found this one thanks to Sacasas. It’s from David M. Kaplan, Ricoeur’s Critical Theory (2003), which I have not yet had the chance to read:

“Technologies are best seen as systems that combine technique and activities with implements and artifacts, within a social context of organization in which the technologies are developed, employed, and administered. They alter patterns of human activity and institutions by making worlds that shape our culture and our environment. If technology consists of not only tools, implements, and artifacts, but also whole networks of social relations that structure, limit, and enable social life, then we can say that a circle exists between humanity and technology, each shaping and affecting the other. Technologies are fashioned to reflect and extend human interests, activities, and social arrangements, which are, in turn, conditioned, structured, and transformed by technological systems.”

I liked Michael’s comment on this beefy definition: “This definitional bloat is a symptom of the technological complexity of modern societies. It is also a consequence of our growing awareness of the significance of what we make.”

Jacques Ellul

Jacques Ellul, a French theologian and sociologist, penned a massive, 440-plus page work of technological criticism in 1954, La Technique ou L’enjeu du Siècle (1954), which was later translated in English as, The Technological Society (New York: Vintage Books, 1964). In setting forth his critique of modern technological society, he used the term “technique” repeatedly and contrasted with “technology.” He defined technique as follows:

“The term technique, as I use it, does not mean machines, technology, or this or that procedure for attaining an end. In our technological society, technique is the totality of methods rationally arrived at and having absolute efficiency (for a given state of development) in every field of human activity. […] Technique is not an isolated fact in society (as the term technology would lead us to believe) but is related to every factor in the life of modern man; it affects social facts as well as all others. Thus technique itself is a sociological phenomenon…” (p. xxvi, emphasis in original.)

Bernard Stiegler

In  La technique et le temps, 1: La faute d’Épiméthée, or translated, Technics and Time, 1: The Fault of Epimetheus (1998), French philosopher Bernard Stiegler defines technology as:

“the pursuit of life by means other than life”

[I found that one here.]

Peter Thiel

In Zero to One: Notes on How to Build the Future (2014), Internet entrepreneur and venture capitalist Peter Thiel says,

“Properly understood, any new and better way of doing things is technology.”

Marc Andressen

Marc Andreessen is interviewed in June 2020 by Sriram Krishan in his newsletter, The Observer Effect, and asked what motivates him to support technological innovation. He closes by defining technology as follows:

“Technology is quite literally the lever for being able to take natural resources and able to make something better out of them.”

Frederick Ferré

Frederick Ferré’s Philosophy Of Technology (1988) is a wonderful introduction to the study of this subject and has become a widely assigned textbook used in many college courses. In Chapter 2, “Defining Technology,” Ferré provided a remarkably concise definition of “technologies” as:

“practical implementations of intelligence” (with the caveat that “‘Practical’ requires that they not be wholly ends in themselves; ‘implementations’ entails that a technology be somehow concretely embodied, normally in implements or artifacts, sometimes simply in social organization…”)

Importantly, Ferré arrived at this definition by carefully detailing what should and should not be considered “technological.” In an attempt to avoid excessive breadth when defining the term, Ferré made four important stipulations:

1. Technology is implemented, not ’empty-handed’: “[I]t would be wise to resist a definition of technology that includes empty hands as technological implements. The totally naked human body, interacting face-to-face with the environment, unmediated by any artifact, contrivance, invention, or tool, would seem to stand as a paradigm case of the non-technological.”
2. Technology is practical, not ‘for its own sake’: Where “the notion of the ‘practical’. . . [means] supporting such ends as survival, health, comfort, and material well-being.”
3. Technology is embodied, non ‘in the head’ alone: “[I]t would be wise to guard against the absorption of all methods and techniques, including wholly mental ones, into the concept of technology.” He uses the examples of natural language and mathematics.
4. Technology is intelligent, not ‘blind’: “[T]he concept of technology will not usefully be extended to behavior that, among humans, is merely accidental or, among other species, is entirely instinctive. . . . Put positively, it suggests our definition will need to stipulate that technology involves (i) implements used as (ii) means to practical ends that are somehow (iii) manifested in the material world as (iv) expressions of intelligence.”

John Fernald

Compared to philosophers, historians, and social critics, economists tend to define technology in a somewhat more dry fashion. (No surprise there, right?!) That being said, it is surprising how few economists bother defining the term in their articles and textbooks. But here’s a concise definition of the term that I recently heard John Fernald, an economist and Senior Research Adviser at the Federal Reserve Bank of San Francisco, articulate at a policy conference. In an October 2014 presentation entitled, “Technology and the American Economy: Or, What’s the New Normal?,” Fernald defined technology as the:

“Ability to convert society’s resources (labor and capital) into output (goods and services that we value).”

Ian Barbour

In Chapter 1 of his 1993 book, Ethics in an Age of Technology, Ian Barbour discussed three  conflicting views of technology: “Technology as Liberator,” “Technology as Threat,” and “Technology as Instrument of Power.” Before discussing each, he defined technology as follows:

“Technology may be defined as the application of organized knowledge to practical tasks by ordered systems of people and machines.” (p. 3)

He continued on to note that:

“There are several advantages to such a broad definition. ‘Organized knowledge’ allows us to include technologies based on practical experience and invention as well as those based on scientific theories. The ‘practical tasks’ can include both the production of material goods (in industry and agriculture, for instance) and the provision of services (by computers, communications media, and biotechnologies, among others). Reference to ‘ordered systems of people and machines’ directs attention to social institutions as well as to the hardware of technology. The breadth of the definition also reminds us that there are major differences among technologies.” (p. 3-4)

Robert Friedel

In his 2007 book, A Culture of Improvement: Technology and the Western Millennium, University of Maryland historian Robert Friedel offers a formal definition of technology to kick off the book and then ends with a less formal one:

“By technology we typically mean the knowledge and instruments that humans use to accomplish the purposes of life.” (p. 1)

He also clarifies the definition by explaining what it does  not include, namely: “processes that completely mental or biological;” “knowledge of the world … that is purely in the realm of ideas and description;” and “nature.”  He then closes the book by noting that:

“Technology can, indeed, be defined as a pursuit of power over nature.” (p. 543).

Again, please feel free to suggest additions to this compendium that future students and scholars might find useful. I hope that this can become a resource to them.

Additional Reading:

• Frederick Ferré – Philosophy Of Technology (1988)
• Carl Mitcham, “From Philosophy to Technology, ” Ch. 6 in Thinking Through Technology: The Path Between Engineering and Philosophy (Chicago: The University of Chicago Press, 1994), 137-160.
• Eric Schatzberg – “Technik Comes to America: Changing Meanings of Technology,” Technology and Culture (2006)

What follows is a response to Michael Sacasas, who recently posted an interesting short essay on his blog The Frailest Thing, entitled, “10 Points of Unsolicited Advice for Tech Writers.” As with everything Michael writes, it is very much worth reading and offers a great deal of useful advice about how to be a more thoughtful tech writer. Even though I occasionally find myself disagreeing with Michael’s perspectives, I always learn a great deal from his writing and appreciate the tone and approach he uses in all his work. Anyway, you’ll need to bounce over to his site and read his essay first before my response will make sense.

Michael:

Lots of good advice here. I think tech scholars and pundits of all dispositions would be wise to follow your recommendations. But let me offer some friendly pushback on points #2 & #10, because I spend much of my time thinking and writing about those very things.

In those two recommendations you say that those who write about technology “[should] not cite apparent historical parallels to contemporary concerns about technology as if they invalidated those concerns. That people before us experienced similar problems does not mean that they magically cease being problems today.” And you also warn “That people eventually acclimate to changes precipitated by the advent of a new technology does not prove that the changes were inconsequential or benign.”

I think these two recommendations are born of a certain frustration with the tenor of much modern technology writing; the sort of Pollyanna-ish writing that too casually dismisses legitimate concerns about the technological disruptions and usually ends with the insulting phrase, “just get over it.” Such writing and punditry is rarely helpful, and you and others have rightly pointed out the deficiencies in that approach.

That being said, I believe it would be highly unfortunate to dismiss any inquiry into the nature of individual and societal acclimation to technological change. Because adaptation obviously does happen! Certainly there must be much we can learn from it. In particular, what I hope to better understand is the process by which we humans have again and again figured out how to assimilate new technologies into their lives despite how much those technologies “unsettled” well-established personal, social, cultural, and legal norms.

To be clear, I entirely agree with your admonition: “That people eventually acclimate to changes precipitated by the advent of a new technology does not prove that the changes were inconsequential or benign.” But, again, we can agree at least agree that such acclimation has happened regularly throughout human history, right?  What were the mechanics of that process? As social norms, personal habits, and human relationships were disrupted, what helped us muddle through and find a way of coping with new technologies? Likewise, as existing markets and business models were disrupted, how were new ones formulated in response to the given technological disruption? Finally, how did legal norms and institutions adjust to those same changes?

I know you agree that these questions are worthy of exploration, but I suppose where we might part ways is over the question of the metrics by which judge whether “the changes were inconsequential or benign.” Because I believe that while technological change often brings sweeping and quite consequential change, there is a value in the very act of living through it.

In my work, including my latest little book, I argue that humans have exhibited the uncanny ability to adapt to changes in their environment, bounce back from adversity, and learn to be resilient over time. A great deal of wisdom is born of experience, including experiences that involve risk and the possibility of occasional mistakes and failures while both developing new technologies and learning how to live with them. I believe it wise to continue to be open to new forms of innovation and technological change, however, not only because it provides breathing space for future entrepreneurialism and invention, but also because it provides an opportunity to see how societal attitudes toward new technologies evolve — and to learn from it. More often than not, I argue, citizens have found ways to adapt to technological change by employing a variety of coping mechanisms, new norms, or other creative fixes.

Even if you don’t agree with all of that, again, I would think you would find great value in studying the process by which such adaptation happens. And then we could argue about whether it was all really worth it! Alas, at the end of the day, it may be that we won’t be able to even agree on a standard by which to make that judgment and will instead have to settle for a rough truce about what history has to teach us that might be summed up by the phrase: “something gained, something lost.”

With all this in mind, let me suggest this friendly reformulation of your second recommendation: Tech writers should not cite apparent historical parallels to contemporary concerns about technology as if they invalidated those concerns. That people before us experienced similar problems does not mean that they magically cease being problems today. But how people and institutions learned to cope with those concerns is worthy of serious investigation. And what we learned from living through that process may be valuable in its own right.

I have been trying to sketch out an essay on all this entitled, “Muddling Through: Toward a Theory of Societal Adaptation to Disruptive Technologies.” [ update: Here it is!] I am borrowing that phrase (“muddling through”) from Joel Garreau, who used it in his book “Radical Evolution” when describing a third way of viewing humanity’s response to technological change. After discussing the “Heaven” (optimistic) and “Hell” (skeptical or pessimistic) scenarios cast about by countless tech writers throughout history, Garreau outlines a third, and more pragmatic “Prevail” option, which views history “as a remarkably effective paean to the power of humans to muddle through extraordinary circumstances.” That pretty much sums up my own perspective on things, but much study remains to be done on how that very messy process of “muddling through” works and whether we are left better off as a result. I remain optimistic that we do!

As always, I look forward to our continuing dialog over these interesting issues and I wish you all the best.

Cheers,

Adam Thierer

[UPDATE 4/30/13: This article was subsequently published in Volume 65, Issues 2 of the Federal Communications Law Journal in April 2013. The links below now point to the final FCLJ version.]

The Mercatus Center at George Mason University has just released a new paper by Brent Skorup and me entitled, “Uncreative Destruction: The War on Vertical Integration in the Information Economy.”  Brent, who is the research director for the Information Economy Project at the George Mason University School of Law, and I have been working on this paper since the Spring and we are looking forward to getting it published in a law review shortly. The paper focuses on Tim Wu’s “separations principle” for the digital economy, something I’ve spent some time critiquing here in the past. Here’s the introduction from the 44-page paper that Brent and I just released:

Are information sectors sufficiently different from other sectors of the economy such that more stringent antitrust standards should be applied to them preemptively? Columbia Law School professor Tim Wu responds in the affirmative in his book The Master Switch: The Rise and Fall of Information Empires. Having successfully pushed net-neutrality regulation into the policy spotlight, Wu has turned his attention to what he regards as excessive market concentration and threats to free speech throughout the entire information economy.To support his call for increased antitrust intervention, Wu explains his view of competition in the information economy—a view that deviates substantially from current mainstream antitrust theory. First, Wu contends that “information monopolies” are pervasive in the information economy. Wu’s “monopolists” include Facebook, Apple, Google, and even Twitter. In The Master Switch and essays like “In the Grip of the New Monopolists,” Wu argues that these so-called monopolies are increasing their market power and require more aggressive oversight and regulation.Second, Wu argues that traditional antitrust analysis is not sufficient for information systems because they carry speech. He claims, “Information industries… can never be properly understood as ‘normal’ industries,”and traditional forms of regulation, including antitrust enforcement, “are clearly inadequate for the regulation of information industries.”Wu believes that because information industries “traffic in forms of individual expression” and are “fundamental to democracy,” they should be subject to greater regulatory treatment.Third, in contrast to current competition law’s focus on horizontal relationships, Wu desires a reinvigorated regulatory enforcement that addresses “the corrupting effects of vertically integrated power” in the information sectors.He is particularly concerned about private threats to free speech arising from such vertical integration.The solution, he says, is preventing vertical mergers in the information economy and the mandatory divestiture of vertically integrated companies. To implement this, Wu proposes a Separations Principle for the information economy, which would segregate information providers into three buckets, which we have labeled information creators, information distributors, and hardware makers.This article outlines Wu’s separations proposal, explains why his fears regarding vertical relationships should be rejected by regulatory and antitrust policymakers, and illustrates the legal and practical problems his Separations Principle poses. Wu justifies his Separations Principle by citing monopolies and market power in the information economy. He also advocates using U.S. antitrust authorities to enforce his Principle. We argue that the antitrust harms he fears are not present, and we highlight scholarship on the accepted benefits of vertically integrated firms. We show that Wu’s remedies are policy preferences wrapped in the language of competition law. In fact, the information economy is largely competitive and does not warrant interventionist regulatory enforcement. Since much of American economic vitality flows from the information economy and technology, policymakers should reject a radical antitrust remedy like Wu’s preemptive Separations Principle.

The paper can be downloaded from the Mercatus website, SSRN, or Scribd.

In this new Money Morning article, “The Antitrust Curse: What Apple Can Learn From Microsoft, IBM,”  David Zeiler wonders whether the antitrust lawsuit filed against Apple and several book publishers by the U.S. Department of Justice last week could open the door to a broader case against Apple or, at a minimum, simply become a major distraction to the firm and it’s ability to innovate going forward. He uses IBM and Microsoft as case studies in this regard and notes that, “the problem with being in the DOJ’s gunsight is that it distracts management, makes the company hesitant to innovate, and blemishes the company’s public image.  While antitrust woes may not have been entirely responsible for Microsoft and IBM ceding their dominant positions in tech, they were clearly a major factor,” he says. “And worse for Apple, the e-book case could be just the beginning.”

Quite right. I raised the same concern in my recent Forbes column,”Regulatory, Antitrust and Disruptive Risks Threaten Apple’s Empire,” which Zeiler was kind enough to quote in his essay. In that piece, I argued:

Even if Apple beats back [the eBooks] investigation, broader questions are being raised about the company’s power that could invite a much broader investigation. The danger for Apple is that antitrust becomes an omnipresent threat that must be factored into all ongoing business decisions. Antitrust is a particular danger to Apple because the firm is highly vertically integrated and that integration is the source of many of their innovations.  As earlier tech titans like IBM and Microsoft learned, when antitrust hangs like the Sword of Damocles, every decision about how to evolve and innovate becomes a calculated gamble.

Regarding the earlier impact that antitrust Sword of Damocles had on Microsoft, Zeiler unearthed this terrific 2005 quote from Mark Kroese, a general manager of information services at the Microsoft Network, who described the impact of the MS antitrust case on innovation at the firm as follows: “Working at Microsoft today vs. five years ago is different,” Kroese said. “If anyone thinks the antitrust case hasn’t slowed us down, you’re wrong. If I want to meet with a products manager for Windows, there needs to be three lawyers in the room. We have to be so careful, we err on the side of caution. We are on such a fine line of conduct.” Regarding how antitrust chilled IBM, Zeiler cites veteran tech journalist Steve Wildstrom of Tech.pinions who noted,  “Twelve years of litigation were an enormous distraction in a time of rapid technological and business change. IBM management became cautious and over-lawyered, constantly looking over its shoulder-a condition that persisted for years after the case ended. The antitrust case was almost certainly a major cause of the serious decline of IBM in the late 1980s and early 90s,” Wildstrom said.

Of course, it is impossible to scientifically determine to what degree antitrust harassment contributed to either IBM or Microsoft’s inability to innovate and adapt to the rapidly changing market conditions. And let’s be clear: both IBM and MS have found ways to rebound and innovate in other ways. But one wonders what was lost in the process as the threat of antitrust constantly loomed and potentially chilled innovative efforts that could have kept both firms on the cutting-edge.

It’s not just Apple that faces similar threats today. Google is obviously another company increasingly mentioned as an antitrust target. Commenting of the dangers of a potential case against Google, Bernstein Research senior analyst Carlos Kirjner argues that “even if regulatory proceedings come to naught, the process has the potential, in the most extreme circumstances, to consume so much of the company’s energy that it can lead to important strategic missteps: many believe that Microsoft missed the boat on the Internet, and IBM on the importance of the personal computer, in large part because their management teams were focused on defending against the DoJ’s antitrust efforts.”

The better approach to disciplining tech firms and markets is to rely less on intervention and more on Schumpeter’s “perennial gales of creative destruction,” which are blowing harder than ever in our modern high-tech economy. In markets built largely upon binary code and governed by Moore’s Law, the pace and nature of change has become hyper-Schumpeterian: unrelenting and utterly unpredictable. Innovative risk-takers are constantly shaking things up and displacing yesterday’s lumbering, lethargic giants. Just ask some of the players that have been largely left in the dust, including AOL, AltaVista, MySpace, Palm, and others. Of course, there’s my favorite recent case study: Research In Motion’s BlackBerry smartphone.  As I noted in my recent column, “Bye Bye BlackBerry. How Long Will Apple Last?” BlackBerry was virtually synonymous with “smartphones” and was considered one of the tech titans that seemed destined to dominate for many years to come. But now the BlackBerry’s days appear numbered and its parent company Research In Motion Ltd. is struggling for its very survival.

Too many tech industry pundits today ignore these dynamic realities and instead rely a myopic analytical approach to the information economy that is fundamentally static in character. Many static equilibrium scholars in both the legal and economic profession tend to adopt a snapshot view of markets and innovation. Such critics often express an overly nostalgic view of the technological past while adopting an excessively gloomy view of the present and the chances for future progress.

But, a la Schumpeter, modern tech markets are highly dynamic. There is no static end-state, “perfect competition,” or “market equilibrium” in today’s information technology marketplace. Change and innovation are chaotic, non-linear, and paradigm-shattering. Schumpeter said it best long ago when he noted how, “in capitalist reality as distinguished from its textbook picture, it is not [perfect] competition which counts but the competition from the new commodity, the new technology, the new source of supply, the new type of organization… competition which commands a decisive cost or quality advantage and which strikes not at the margins of the profits and the outputs of the existing firms but at their foundations and their very lives. This kind of competition is as much more effective than the other,” he argued, because the “ever-present threat” of dynamic, disruptive change “disciplines before it attacks.”

By contrast, the static equilibrium mindset is myopically fixated on short-term market share and price competition while ignoring “competition for innovation,” which is what matters most in the more dynamic Schumpeterian model. “Schumpeterian competition is primarily about active, risk-taking decision makers who seek to change their parameters,” note economists Jerry Ellig and Daniel Lin. “It is about continually destroying the old economic structure from within and replacing it with a new one.” Thus, while static or “perfect competition” models assume away innovation and are preoccupied with equilibrium, dynamic models revolve around disequilibrium and assume that the only constant is change. What is most important to economic progress, therefore, is the ongoing process of constant experimentation and spontaneous discovery that allows new business models and organizational structures to emerge in response to market signals.

The other danger of the static equilibrium mindset is that the same new innovators and innovations that obtain success and scale quite rapidly as a result of this process are sometimes thought to possess problematic market power. Accusations of “monopoly” quickly follow. As Nobel Laureate Ronald Coase noted, “if an economist finds something—a business practice of one sort or another—that he does not understand, he looks for a monopoly explanation. And as in this field we are very ignorant, the number of understandable practices tends to be very large, and the reliance on a monopoly explanation, frequent,” he argued.  Of course, non-economists are just as likely—perhaps more likely—to make that same error. This is why a short-term fixation on market share and market power is so problematic.

Moreover, as Schumpeter also taught us, it is essential that uneven entrepreneurial gains be tolerated so that innovation can occur and be continuously incentivized. Economies need innovators to take risks because progress is born from it. Penalizing the risk-takers by trying to “level the playing field” through rash regulation or antitrust interventions will simply sap the entrepreneurial spirit from the marketplace, limit technological innovation, and diminish the possibility of progress and prosperity over the long-haul.

If you’d like a better understanding of this dynamic conception of competition and an explanation of why the static equilibrium mindset — especially in the antitrust field — is so horribly misguided, then I strongly recommend you begin your investigation with the following readings:

Over at Twitter, our TLF blogging colleague Jerry Brito asks a smart question about the Federal Communications Commission’s recently-opened investigation of the Apple-Google spat over Apple’s recent decision to reject the Google Voice app from the iPhone App Store.  Jerry asks: “Maybe I should know this, but what authority does the FCC have to demand that Apple explain anything?”  Good question, Jerry!  But no, I don’t think there’s anything you’re missing.  We might consider this merely the latest chapter of the agency’s rogue operator history: If you can’t find the authority to do something, just assert it anyway and go for broke!  The idea of living within the confines of the law and paying attention to statutory authority seems like an alien concept to the FCC.  As my PFF colleagues Barbara Esbin and Adam Marcus have pointed out in their amazing recent law review article, “The Law Is Whatever the Nobles Do: Undue Process at the FCC,” when all else fails, the agency just asserts “ancillary jurisdiction” and claims that the whole world is their oyster. They argue:

The FCC’s means of asserting regulatory authority over broadband Internet service providers’ (“ISP”) network management practices is unprecedented, sweeping in its breadth, and seemingly unbounded by conventional rules of interpretation and procedure. We should all be concerned, for apparently what we have on our hands is a runaway agency, unconstrained in its vision of its powers.

Of course, even if we ignore the agency’s cavalier attitude about the law and statutory authority, there are other reasons to be concerned about FCC interference in this matter. Berin and Ryan have already pointed out the other side of the story: That this is just old-fashion cut-throat competition, and that consumers continue to enjoy rapidly expanding options in this marketplace. [Also see this paper that Barbara Esbin and Berin co-authored: Should the FCC Kill The Goose That Laid The Golden iPhone.] And even some of those folks in the press or the blogosphere who welcome some FCC oversight in this case recognize the horrific potential downside here.  As Larry Dignan, Editor in Chief of ZDNet, argues in his piece today, “FCC’s More Proactive Stance: Should We Cheer or Worry?”:

But then there’s the other side of the equation. The one that can make you squirm. The FCC is looking into everything from app approval to exclusive deals between carriers and device makers. At some point, the FCC meddles in free markets. It will micromanage. For now, the FCC’s moves require a wait-and-see approach, but it’s clear there’s a new sheriff in town and he isn’t going to be shy about probing all aspects of the wireless business. Stay tuned to see how this turns out.

Uh, yeah. And that’s what has some of us so worried. When the FCC “meddles” and “micromanages” the results are usually less than stellar.  Once the FCC starts regulating every aspect of our smartphones, chances are they won’t be so smart any more.  In just one year’s time, the Apple iPhone Store has facilitated 1.5 Billion downloads of over 65,000 free and paid apps by consumers in 77 countries. Does anyone think the FCC is going to do better than that once they start micro-managing the process?