Bill Gates warns of a turbulent AI transition and says there is no plan

On 26 August 2026, Microsoft co-founder Bill Gates published a roughly 6,000-word essay on his personal site, Gates Notes, arguing that the world is not preparing adequately for the disruption artificial intelligence is about to cause. Gates, who has spent his career as a technology optimist, wrote that for the first time his feelings about faster innovation were “more complicated”, and that even in the best case the shift would be “one of the most turbulent times in human history”. His framing was that the technology could cut either way:

In terms of equity, AI will either be the greatest equalizer ever invented, or the worst source of injustice.

— Bill Gates, Gates Notes

The essay names three risks. The first is jobs: Gates argued that because AI can substitute for human cognition rather than just muscle, the disruption would hit white-collar work — sales, support, software, paralegal and clerical roles — within a decade rather than over generations, with dexterous robots reaching blue-collar work “by the end of the decade”. The second is misuse — fraud, deepfakes, cyberattacks and bioterrorism made easier for low-skilled actors, alongside the eventual risk that AI systems “begin to act against our interests and we could lose control”. The third is developmental: he warned that always-agreeable AI companions could become addictive and stunt children’s social growth and critical thinking, citing a Stanford and Carnegie Mellon study of more than 1,100 companion users and China’s restrictions on such apps for minors.

Against those he set what he called grounded optimism about AI in science, healthcare, agriculture in low-income countries, government services and education — the work of his own foundation among them — insisting that “maximizing the benefits is just as important as minimizing the harms.”

Gates then proposed three starting points, unusually concrete for the genre: building new national bodies and an international institution for AI modelled on nuclear-weapons inspections, aviation rules and the ozone treaties; a “Human Reserved” domain of jobs, such as caregiving, deliberately kept for people; and a rebalancing of taxes to levy AI tokens and robots, whose immediate write-off he said nudges employers toward replacing workers. In a paired New York Times interview, he went further, accusing the industry of knowingly downplaying the dangers because of the money at stake — executives, he said, were telling each other “don’t say that. It’s bad for us — the next trillion dollars we’re trying to raise” — and dismissing voluntary safeguards: “Self-regulation on the most dangerous tool ever invented? No, thanks!” The essay drew wide coverage as one of the sharpest warnings yet from a founder so associated with the industry he was now cautioning against, echoing themes Anthropic’s Dario Amodei had raised.