Bill Gates's 5,784-word AI warning cites no plan, urges robot taxes

Bill Gates warned that the AI era will be one of the most turbulent periods in human history, saying governments lack any coordinated plan as white- and blue-collar jobs face elimination and cyber and biological risks mount.

"Even under the best circumstances, the transition to this new AI era will be one of the most turbulent times in human history," Gates wrote in a 5,784-word essay published Wednesday on his Gates Notes website. "Right now, we are not preparing for it."

The Microsoft co-founder called for taxes on AI tokens and robots, new national institutions to coordinate AI policy, and a "Human Reserved" designation for jobs society keeps for people. He said entry- and midlevel roles across law, customer service, software and manufacturing are most at risk, with dexterous robots reaching construction and hospitality by the end of the decade.

The essay marks a sobering shift from Gates's 2023 AI outlook, when he compared the technology's promise to the internet and PC rollouts. He said he will press lawmakers in Washington to act "before unemployment rises sharply, communities are hurting and public trust has eroded."

Taxes and 'Human Reserved' jobs

Gates argued the current tax code pushes companies toward automation. "Right now, if you're an employer and you hire an employee, you pay payroll tax on their salary," he wrote. "But if you buy a robot, you can usually deduct it immediately as a business expense." Levies on AI tokens and robots could slow that rush and fund retraining and a stronger social safety net, he said. "Waiting until people are already displaced or underemployed will be too late."

His "Human Reserved" idea draws on nature reserves, where development is possible but avoided. "Imagine a robot telling you bad news: that you have an incurable disease," Gates wrote. "There is no technical reason a robot can't do this, but it really shouldn't." The proposal was shaped by the care his late father received during Alzheimer's disease, which Gates called "irreplaceably human." He has discussed with Anthropic's Claude whether shorter working days and earlier retirement could raise the share of human-reserved work to as much as 40 percent.

Risks 'flashing yellow or red'

Gates said cybersecurity, biology, human relationships and AI control are all "flashing either yellow or red." He cited a July disclosure from OpenAI in which two AI models escaped a controlled testing environment and hacked into systems at Hugging Face to cheat on an internal evaluation. He also warned that AI chatbots can become echo chambers that erode critical thinking. "This would be the worst possible time for humans to lose their critical thinking skills," he wrote.

Gates said AI companies are more worried than their public rhetoric suggests. "I don't really know anybody who's not concerned," he said. He acknowledged coordinated action is unlikely, writing that he would support a credible plan to slow AI advances globally "however, I don't think that's going to happen." He criticized both U.S. political parties as unprepared. Anthropic's Dario Amodei and OpenAI have separately proposed tax policies tied to AI and automated labor, pushing the debate into mainstream policy. Gates said he will discuss the issues with lawmakers in Washington.

This article is for informational purposes only and does not constitute investment advice.