Microsoft founder Bill Gates warned of the dangers of artificial intelligence and directly rejected the industry self-regulation approach championed by U.S. President Donald Trump.
Speaking on the New York Times podcast "The Ezra Klein Show" on the 30th, Gates said the industry cannot be trusted to police itself on AI, calling the idea "really insane."
Asked about the argument that no separate regulation is needed because companies that release unsafe AI will be sued, Gates said he could hardly believe the question was being raised, adding that AI is the most dangerous thing humanity has ever faced.
He pointed to other industries to expose the gaps in the self-regulation argument. The situation, he said, is no different from having no Food and Drug Administration to review medicines, no agency overseeing aviation safety and no seat belt mandate for cars. As for the logic of using lawsuits to prevent harm, he asked whether that meant waiting until someone had killed 100 million people with a biological weapon and then filing suit. Gates said advances in AI have already created conditions in which malicious actors could carry out cyberattacks or biological terrorism.
He assessed awareness of AI risks among think tanks, politicians and policymakers as very low. He suggested one reason may be that rockets and nuclear weapons were built largely with government research and development money, while AI has been developed with private capital. He urged that safeguards and oversight systems be made mandatory before a major disaster occurs.
As a remedy for the shock to the labor market, he proposed a "robot tax." The idea is to apply the social security taxes levied on hiring people to robots and AI as well, slowing the pace at which jobs are replaced and using the revenue to retrain displaced workers and expand the social safety net.
Gates also floated the idea of designating "human-protected jobs," areas where AI would be deliberately kept out, much like nature preserves. The aim is to guarantee human employment in fields that require empathy and emotional connection, such as child care, elderly care, health care and education. In low-income countries, however, he said AI could serve as a tool to fill infrastructure gaps in health care and agriculture, stressing that technical support for those countries should be accelerated.
Gates also addressed concerns that AI could bring about the extinction of humanity. "If humans keep control, we will eventually enter an era of superabundance," he said, adding that the transition must be navigated wisely.
In a recent interview with NBC, he also urged the U.S. government to tighten regulation, saying that AI in the hands of malicious people could trigger an event killing as many as 1 billion people.
A day earlier, Trump dismissed federal regulation and signed an industry-led "self-regulation" agreement. After a luncheon with chief executives of major U.S. AI companies at the White House, Trump said there has to be "a tremendous amount of self-regulation," adding that while regulation is happening automatically through the Justice Department and the Federal Bureau of Investigation, self-regulation is very important.
