Does it really come down to balancing innovation with safety? Or is the right way forward more complex?

This week’s Executive Order on Safe, Secure, and Trustworthy Artificial Intelligence marks an important advancement toward much-needed regulation of the development and application of AI. The latest generation of generative AI models, most notably OpenAI’s GPT-4 and ChatGPT, their general use chatbot, have already profoundly affected how millions of people live and work, worldwide. 

An overview of the 111-page document

The Executive Order seeks to “advance and govern the development and use of AI in accordance with eight guiding principles and priorities.” Taken from the full text of the Executive Order, they are:

1. Artificial Intelligence must be safe and secure. Meeting this goal requires robust, reliable, repeatable, and standardized evaluations of AI systems, as well as policies, institutions, and, as appropriate, other mechanisms to test, understand, and mitigate risks from these systems before they are put to use.  
2. Promoting responsible innovation, competition, and collaboration will allow the United States to lead in AI and unlock the technology’s potential to solve some of society’s most difficult challenges. This effort requires investments in AI-related education, training, development, research, and capacity, while simultaneously tackling novel intellectual property (IP) questions and other problems to protect inventors and creators. 
3. The responsible development and use of AI require a commitment to supporting American workers. As AI creates new jobs and industries, all workers need a seat at the table, including through collective bargaining, to ensure that they benefit from these opportunities. My Administration will seek to adapt job training and education to support a diverse workforce and help provide access to opportunities that AI creates.
4. Artificial Intelligence policies must be consistent with my Administration’s dedication to advancing equity and civil rights. My Administration cannot—and will not—tolerate the use of AI to disadvantage those who are already too often denied equal opportunity and justice. From hiring to housing to healthcare, we have seen what happens when AI use deepens discrimination and bias, rather than improving quality of life. 
5. The interests of Americans who increasingly use, interact with, or purchase AI and AI-enabled products in their daily lives must be protected. Use of new technologies, such as AI, does not excuse organizations from their legal obligations, and hard-won consumer protections are more important than ever in moments of technological change. The Federal Government will enforce existing consumer protection laws and principles and enact appropriate safeguards against fraud, unintended bias, discrimination, infringements on privacy, and other harms from AI. 
6. Americans’ privacy and civil liberties must be protected as AI continues advancing. Artificial Intelligence is making it easier to extract, re-identify, link, infer, and act on sensitive information about people’s identities, locations, habits, and desires. Artificial Intelligence’s capabilities in these areas can increase the risk that personal data could be exploited and exposed.   
7. It is important to manage the risks from the Federal Government’s own use of AI and increase its internal capacity to regulate, govern, and support responsible use of AI to deliver better results for Americans. These efforts start with people, our Nation’s greatest asset. My Administration will take steps to attract, retain, and develop public service-oriented AI professionals, including from underserved communities, across disciplines—including technology, policy, managerial, procurement, regulatory, ethical, governance, and legal fields—and ease AI professionals’ path into the Federal Government to help harness and govern AI.  
8. The Federal Government should lead the way to global societal, economic, and technological progress, as the United States has in previous eras of disruptive innovation and change. This leadership is not measured solely by the technological advancements our country makes.  Effective leadership also means pioneering those systems and safeguards needed to deploy technology responsibly—and building and promoting those safeguards with the rest of the world.  

Not all AI applications are created equal

These eight principles do an excellent job of implicitly distinguishing between two different types of AI applications. The first is where AI is introduced to supplant human decision-making, often seen in policing, sentencing, education admission, and housing decisions. Because these uses significantly impact people’s lives and can introduce a high risk of bias, they require timely and targeted controls, and the White House is appropriately aggressive on these. This is the higher risk of the two applications, and the principles identify appropriate controls for this category.

The second  set of applications use AI to elevate information humans (and entire professions) can use to drive better, often more efficient decision-making. These are the applications where accuracy and transparency is key, but the risks are quite different and more limited, and thus the emphasis is, appropriately, on incentivizing and promoting innovation.

Few professions have been or will be more affected than law. Given what we’ve heard from so many CoCounsel customers about how much our AI legal assistant has improved their day-to-day work and the services they deliver to their clients, we’re encouraged that the Executive Order balances the real need for controlling fraudulent, misleading, or inegalitarian uses of AI with the once-in-a-lifetime opportunity to promote applications that can drive unprecedented value and benefit to Americans across so many facets of our lives. The potential for legal applications built on top of LLMs to advance access to justice in America, where more than 80% of all civil legal disputes are not represented by counsel, cannot be overstated. 

A dynamic AI ecosystem, subject to necessary accuracy, safety, and privacy controls, can certainly and has already improved the legal profession, but it also has the potential to do even more to expand access to justice, address climate change, and tackle a range of other societal challenges with greater impact than we ever thought possible.

The best solutions lie somewhere in the middle

These gains will only come, however, if informed concerns are thoroughly explored, such as whether the proposed actions will “stifle” innovation due to “overhyped dangers of AI,” or whether that innovation will make us “more fair, just, and prosperous, rather than surveilled, silenced, and stereotyped.” Putting regulations in place does not have to stifle innovation. At the same time, to believe that AI innovation should not be slowed down at all is unrealistic, and disingenuous. The pace of unchecked innovation cannot be our benchmark. Regulated innovation will be slower, but the alternative is unacceptable.

While it is important to enact regulation that will still permit us to move as quickly as possible, it’s critical that new standards and programs are informed by subject matter experts who deeply understand not just the underlying technology but specific applications, opportunities, and risks. Instead of being boiled down to “safety vs innovation,” there must be open and ongoing dialogue among the White House, Congressional leaders, developers of foundational models, innovators of new applications, and the communities impacted. This dialogue is what should inform controls and policies that both are tailored to specific risks and promote world-positive applications. 

It’s vital that these measures deeply consider issues that exist in the vast grey area between two false extremes. Because doing right by the opportunities for AI and the people affected by it—for better and for worse—requires thoughtfulness, not soundbites.