Bill Gates is taking a much more alarmed position on AI, arguing that some technology executives privately recognize how quickly the technology is improving but are reluctant to emphasize the risks publicly. According to Gates, the enormous amount of money flowing into AI creates an incentive to keep the message optimistic because stronger warnings could make investors nervous. He says executives are effectively worried about the technology while simultaneously pushing ahead because the financial stakes are enormous.
Gates is particularly concerned about job displacement, cybersecurity and biological risks. He says recent advances in AI coding tools such as Claude Code surprised him because they demonstrated capabilities beyond what he expected from software systems. He believes AI could be different from previous technological revolutions because it can automate work across many industries simultaneously rather than simply eliminating certain tasks while creating comparable numbers of new jobs.
His proposed response is considerably more interventionist than the industry's voluntary approach. Gates has suggested a “token tax” on AI that replaces human workers, using the revenue to support displaced employees, as well as protecting certain “Human Reserved” jobs where automation could cause particularly severe social consequences. He also wants mandatory testing of AI systems capable of assisting with dangerous biological research and stronger international agreements governing those capabilities.
The broader takeaway is that the AI debate is shifting from whether the technology is powerful to whether society can absorb its consequences quickly enough. Gates still believes AI could dramatically improve healthcare, education and living standards, but he argues that innovation alone will not guarantee those benefits are distributed fairly. His warning is particularly significant because it comes from someone who spent decades promoting technological progress: the question is no longer simply how fast can AI advance? but whether governments and institutions can build safeguards fast enough to keep up with it.
