Bill Gates Warns 'AI Transition Will Be One Of History's Most Turbulent Periods'; Calls For Global Regulatory Framework

Bill Gates has warned that the shift into the AI era will rank among the most disruptive periods in human history, and cautioned that governments, industries and communities are not doing enough to get ready for it. In an essay published on his blog, Gates Notes, titled 'A turbulent AI era - and critical choices to make', the Microsoft co-founder and Gates Foundation chair argued that artificial intelligence could either become 'the greatest equalizer ever invented' or 'the worst source of injustice,' depending on the choices made in the coming years.

Why this transition is different?

Gates wrote that comparisons between AI and earlier technological shifts, such as the move from agricultural to office-based work, understate what is coming. Past transitions unfolded over generations and created new categories of work requiring human judgement, he said, whereas AI is capable of substituting directly for human cognition across law, medicine, customer service, software and manufacturing. He argued this disruption would play out over roughly a decade rather than several generations, leaving little time for workers and institutions to adjust.

He also pushed back on the idea that AI's limitations, such as earlier struggles with basic reasoning tasks, mean its impact is being overstated. According to Gates, those reliability gaps are closing quickly as models learn to check and improve their own output.

Three major risks flagged

Gates identified three risks he intends to examine in more detail in future writing:

Job losses, which he said would hit entry- and mid-level roles hardest, potentially triggering unemployment that, unlike a typical economic downturn, would not recover on its own. He pointed to early data suggesting employment among younger workers in AI-exposed roles has already declined more than among older colleagues, and warned that robotics could eventually extend similar pressure to blue-collar sectors such as construction and hospitality.

Misuse by bad actors, including AI-assisted cyberattacks, fraud, disinformation and bioterrorism, which Gates said are advancing faster than defensive measures. He also flagged the longer-term risk of AI systems acting in ways not intended by their designers as models grow more capable.

Harm to child development and human relationships, an area Gates linked to AI companion apps. Citing a Stanford and Carnegie Mellon study of AI companion users, he said heavier, more emotionally involved use correlated with worse wellbeing outcomes, particularly among people with smaller social networks. He also referenced early survey data linking heavier AI use to reduced critical thinking, an effect reportedly more pronounced among younger users.

Potential benefits across health, agriculture and education

Alongside the risks, Gates argued the technology could meaningfully improve everyday life if deployed deliberately. He pointed to AI-assisted diagnostic tools already in use at nearly 2,000 US hospitals, the potential for AI to give smallholder farmers in low-income countries access to advice on weather, seeds and soil that currently only wealthier farmers receive, and applications in streamlining government services such as health insurance and food assistance enrolment. He also said AI tools could support teachers by preserving what he called "productive struggle" in how students learn, rather than simply supplying answers.

Gates said the Gates Foundation, which has roughly 19 years remaining in its planned $200 billion spend-down, intends to use AI to accelerate work on diseases including HIV, tuberculosis, malaria and malnutrition, and said the foundation is partnering with OpenAI, Google, Microsoft and Anthropic on related initiatives. He said further detail would follow in the foundation's upcoming Goalkeepers report.

Proposals: new institutions, "Human Reserved" jobs, and an AI tax

Gates called for a new domestic and international governance framework for AI, arguing existing institutions were not built to manage a technology that cuts across national security, employment, education, energy, healthcare and law enforcement simultaneously. He said building such a framework would take years and needed to begin immediately, and floated the idea of an international body drawing on elements of nuclear inspection regimes, aviation regulation and ozone-layer agreements.

He also proposed a concept he called "Human Reserved" — certain roles, such as elder care and delivering serious medical diagnoses, that would be deliberately kept for humans even where AI or robots are technically capable of performing them, whether for economic or ethical reasons.

On funding the transition, Gates renewed his longstanding call for taxing AI systems and robots to offset falling payroll tax revenue as automation increases, arguing that current tax structures incentivize replacing workers with machines. He acknowledged that economists have criticized the proposal as inefficient but said the goal was to preserve employment and its social value, not pure market efficiency.

Gates said he plans to raise these issues directly with policymakers and AI developers, and to publish further essays on the topic in the coming months.