Should artificial intelligence be regulated?

The creators and developers of ChatGPT believe the new superintelligence could surpass human experts in most disciplines within the next 10 years, and they are calling for supra-regulation, although its chief executive, Sam Altman, does not find the regulation proposed by the European Union acceptable.

The creators of ChatGPT have published a note (which can be seen almost immediately below) warning that within ten years artificial intelligence systems may exceed human experts in most areas. This superintelligence, they say, will be more powerful than other technologies humanity has had to face in the past and poses an existential risk to our civilization. That is why they urge authorities to start thinking now about how we are going to manage it. It is material to examine and, above all, to think about why the creators of the technology are, in theory, so alarmed…

Sam Altman, Greg Brockman and Ilya Sutskever, three of the cofounders of OpenAI, the company behind ChatGPT, believe that the superintelligence of the future will be even more capable than artificial general intelligence — a form of synthetic intelligence comparable to human intelligence — and will be able to be as productive as today's largest companies.

But it is important that we see the content of that note:

The note from the three founders is dated May 22, 2023, and reads verbatim: Responsible artificial intelligence, safety and alignment. Given the picture as we see it now, it's conceivable that within the next ten years, AI systems will exceed expert skill level in most domains, and carry out as much productive activity as one of today's largest corporations. In terms of both potential upsides and downsides, superintelligence will be more powerful than other technologies humanity has had to contend with in the past. We can have a dramatically more prosperous future; but we have to manage risk to get there. Given the possibility of existential risk, we can't just be reactive. Nuclear energy is a commonly used historical example of a technology with this property; synthetic biology is another example. We must also mitigate the risks of today's AI technology, but superintelligence will require special treatment and coordination. A starting point: There are many ideas that matter for us to have a good chance at successfully navigating this development; here we lay out our initial thinking on three of them. First, we need some degree of coordination among the leading development efforts to ensure that the development of superintelligence occurs in a manner that allows us to both maintain safety and help smooth integration of these systems with society. There are many ways this could be implemented: major governments around the world could set up a project that many current efforts become part of, or we could collectively agree (with the backing power of a new organization like the one suggested below) that the rate of growth in AI capability at the frontier is limited to a certain rate per year. And of course, individual companies should be held to an extremely high standard of acting responsibly. Second, we are likely to eventually need something like an IAEA for superintelligence efforts; any effort above a certain capability threshold (or resources like compute) will need to be subject to an international authority that can inspect systems, require audits, test for compliance with safety standards, place restrictions on degrees of deployment and levels of security, etc. Tracking compute and energy usage could go a long way, and give us some hope this idea could actually be implementable. As a first step, companies could voluntarily agree to begin implementing elements of what such an agency might one day require, and as a second, individual countries could implement it. It would be important that such an agency focus on reducing existential risk and not issues that should be left to individual countries, such as defining what an AI should be allowed to say. Third, we need the technical capability to make a superintelligence safe. This is an open research question that we and others are putting a lot of effort into. What's not in scope: We think it's important to allow companies and open-source projects to develop models below a significant capability threshold, without the kind of regulation we describe here (including burdensome mechanisms like licenses or audits). Today's systems will create tremendous value in the world and, while they do have risks, the level of those risks feels commensurate with other Internet technologies and society's likely approaches seem appropriate. By contrast, the systems we are concerned about will have power beyond any technology yet created, and we should be careful not to water down the focus on them by applying similar standards to technology far below this bar. Public input and potential: But the governance of the most powerful systems, as well as decisions regarding their deployment, must have strong public oversight. We believe people around the world should democratically decide on the bounds and defaults for AI systems. We don't yet know how to design such a mechanism, but we plan to experiment with its development. We continue to think that, within these wide bounds, individual users should have a lot of control over how the AI they use behaves. Given the risks and difficulties, it's worth considering why we are building this technology. At OpenAI, we have two fundamental reasons. First, we believe it's going to lead to a much better world than what we can imagine today (we are already seeing early examples of this in areas like education, creative work, and personal productivity). The world faces a lot of problems that we will need much more help to solve; this technology can improve our societies, and the creative ability of everyone to use these new tools is certain to astonish us. The economic growth and increase in quality of life will be astonishing. Second, we believe it would be unintuitively risky and difficult to stop the creation of superintelligence. Because the upsides are so tremendous, the cost to build it decreases each year, the number of actors building it is rapidly increasing, and it's inherently part of the technological path we are on, stopping it would require something like a global surveillance regime, and even that isn't guaranteed to work. So we have to get it right.

According to what we have seen, the vision of OpenAI — a company in which Microsoft participates following its $10 billion investment — is that future generations of artificial intelligence will bring us a much more prosperous future. Although for that to happen, we have to manage the existential risk posed by its implementation.

The creators of ChatGPT think we still have time to manage to develop this superintelligence in a controlled way. Nor do they believe that stopping now is advisable; for them, halting development would be risky and difficult, although it should be remembered that OpenAI's stated purpose is precisely to bring one of these superintelligent systems to life. Besides, there is too much money and too many interests at stake.

The text suggests that the only way to develop this technology safely would be through a global surveillance regime, although they are not sure that would work. That is why they demand that decisions about its development and implementation have “strong public oversight,” something the company's CEO has been repeating at every one of his public appearances.

The problem is how to avoid the existential risk. They offer three ideas for doing so. First, a coordinated development of this technology must be achieved that allows its integration into society, while maintaining some basic safety limits.

To do this they propose two approaches. One, driven by the world's main governments, with the main current developers taking part. And the other, a collective agreement overseen by an independent organization that sets an annual growth rate for the technology. In any case, they note, companies must be required to meet an extremely high level of responsibility.

Second, they propose creating a body similar to the one that regulates atomic energy (the IAEA) to monitor progress toward superintelligence. If a technology exceeds a certain capability threshold, it will have to submit to an audit by an international authority that can inspect the new systems to avoid an existential risk. Something that, they suggest, developer companies could already begin doing voluntarily, as a rehearsal of the capabilities the new body should have.

Finally, the three cofounders of OpenAI assert that the technical capability to make superintelligence safe is needed. “This is an open research question that we and others are putting a lot of effort into,” they say.

But, curiously, OpenAI does not like European laws.

Altman, who besides being a cofounder of OpenAI is also its chief executive, recently warned that his company could stop operating in Europe if the bloc applies its measures on artificial intelligence.

European Justice Commissioner Didier Reynders during a press conference on Artificial Intelligence. The European Union has become the de facto regulator of Artificial Intelligence in the world.

The EU's proposed Artificial Intelligence Act establishes three levels of risk for these technologies. The highest-risk category “would create an unacceptable risk, such as government-managed social scoring of the type used in China” and would be banned. The second risk category would be “subject to specific legal requirements” and would cover AI systems used to scan résumés and rank job applicants. And the third would be for AI systems that “are not explicitly banned or listed as high-risk” and that would therefore “remain largely unregulated.”

The European Parliament voted in favor of the law by a wide majority and set June 14 as the provisional date for its adoption. “The details really matter. We will try to comply, but if we can't do it, we will cease operating,” Altman said in remarks reported by the Financial Times. Unfortunately, it sounds more like a certain threat than anything else.

In tomorrow's edition of the blog we will see what this regulation consists of, which is surely on its way to going into effect in the European Union.