One consistent complaint against the Trump administration is that it is imposing personalist rule, in which the whims of the leader override the law. Now Donald Trump has come out forcefully against regulating artificial intelligence—and his argument is based almost explicitly on the superiority of his omniscience over any formal legal process. He is not so much refuting the charge of personalism as turning it into a promise.
“The only control or ‘guardrails’ that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT, and the U.S.A. has that, in spades!” he wrote in a self-refuting post on Truth Social.
Even the most market-friendly economist will concede that government intervention is required when any private activity creates an “externality”—that is, a cost imposed on people who are not involved in a transaction. A factory dumping toxic waste into a stream has heretofore provided the textbook case of an externality necessitating regulation. But AI bots destroying human civilization—as executives are warning could happen—clearly represents a much costlier externality. This is, in fact, probably the strongest conceivable case for regulation.
The intense competitive pressure among AI firms, which have a much stronger incentive to innovate quickly than to ensure their innovations are safe, enhances the case further.
Skeptics have valid reasons to worry about poorly designed regulation that either entrenches the current industry leaders or permits Chinese AI to overtake the American version. Those are reasons to demand well-designed regulations, not to avoid any rules at all.
[Read: The singularity is not what it seems]
Against this overwhelming rationale, Trump offers, in place of any neutral standards, his characteristic faith in his own prerogatives. Trump tends to discount the need for laws even in areas where he favors government action. He opposed congressional authorization for war with Iran. He has not asked Congress to levy tariffs, preferring instead to raise or lower them and provide carve-outs to affected firms based on his whim.
If anyone needs it, the incumbent president’s behavior provides excellent evidence against personalist rule in general and, in the case of AI, specifically. For one thing, his decisions are not necessarily aligned with the public interest. Trump has formed a political alliance with a handful of wealthy men who have a strong personal stake in the development and shape of AI. The dual roles of David Sacks, Trump’s most important AI adviser and an investor in AI, have caused even some White House officials to express concern about conflicts of interest.
Additionally, Trump has a long-standing tendency to focus on short-term concerns over long-term ones. Trump’s preference for immediate gratification explains many of his decisions, including defunding scientific and medical research, accelerating fossil-fuel emissions, and denigrating American soft power globally. Pushing forward AI development delivers financial benefits to Trump allies who have plied his party with donations. Trump seems to believe that the AI build-out is driving the economy at the moment. It is difficult to imagine any long-term concern persuading Trump to forgo the short-term benefits of the AI boom.
Finally, on the IQ point, Trump is correct that a high level of analytic intelligence is necessary to regulate wisely. This is not, however, a reason to empower Donald Trump. Although he possesses a distinct animal cunning and an uncanny sense for human weakness, Trump isn’t the type to dig in and understand new problems in all of their complexity. He does not appear to read texts of any length or substance, and he is reportedly unable to tolerate briefings of any depth. His public remarks are rambling and error-filled, and rely on a shrinking pool of vocabulary terms.
Trump does have expertise in a handful of realms, such as construction, entertainment, and men’s discount shoeware. Artificial intelligence and technology in general are not among his fields of specialization.
There are many reasons the United States needs clear, stable regulations on artificial intelligence, rather than relying on the whims of a leader. That is true even if the leader is public-minded and deeply versed in policy intricacies. It is all the more true if the leader is Donald Trump.
The post Who Needs Regulation When Trump Is President? appeared first on The Atlantic.
