Bill Gates is deeply worried about AI, and he’s no longer staying quiet
Developed over time first seen 2 months ago
Bill Gates has published a nearly 6,000-word essay warning that artificial intelligence could bring unusually rapid and severe social disruption, marking a sharp shift from his earlier optimism. He argues that the world is unprepared for AI’s effects on employment, inequality and public trust, and calls for governments to act before large-scale job losses take hold. The article says Gates has also linked part of his recent public reticence to his past association with Jeffrey Epstein, which he has called a “huge mistake”.
Gates predicts that AI could permanently eliminate both white-collar and blue-collar jobs, and proposes taxing AI tokens and robots to deter employers from replacing workers and fund social support. He also suggests designating some roles as “Human Reserved”, while urging countries to create bodies able to coordinate AI policy across government and establish a new international organisation for cross-border risks. Although he says cooperation, including between the US and China, is urgent, the article notes that his proposals offer few concrete details on implementation.
- Gates warns AI may cause permanent, widespread job losses.
- He proposes taxing AI and reserving some work for humans.
- He urges new national and global AI governance bodies.
New here? Start with this
Bill Gates co-founded Microsoft and has since become one of the world's best-known philanthropists through the Bill & Melinda Gates Foundation. He has followed the rapid rise of artificial intelligence closely, previously voicing optimism about its potential to solve major global problems such as disease and poverty.
His new essay marks a shift in tone, expressing concern about how quickly AI is changing the world of work and society more broadly. The piece also touches on his past ties to the late financier Jeffrey Epstein, which he has previously acknowledged as an error of judgement.
The story matters because Gates remains an influential figure in technology and public policy, and his changed stance adds a prominent voice to the wider debate about how governments, businesses and international bodies should respond to AI's effects on jobs and society.
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Those sympathetic to Gates argue that a technologist who helped build the digital age is uniquely placed to recognise when a new technology's pace of change is outstripping society's capacity to adapt, and that his warnings deserve weight precisely because he was previously an AI optimist. They contend that automation could plausibly hollow out both white-collar and blue-collar work simultaneously, something no past technological shift has done, making pre-emptive measures such as taxing displacement, ring-fencing certain human-only jobs, and building new international oversight bodies a prudent form of insurance rather than alarmism. On this view, existing national regulators and treaties were not designed for a technology that crosses borders and industries so completely, so genuinely new institutions, including US-China cooperation, are a reasonable response to a genuinely novel risk.
The case against
Sceptics argue that dramatic predictions of mass, permanent job losses have accompanied nearly every major technological shift, from mechanisation to computing, and have consistently underestimated humans' ability to adapt and create new forms of work, so Gates's forecasts should be treated with caution rather than as a basis for policy. They contend that taxing AI tokens and robots, or reserving jobs exclusively for humans, risks slowing genuinely beneficial innovation, entrenching inefficiency, and disadvantaging economies that adopt such measures against those that do not. They further question whether a wealthy tech insider with commercial ties to the industry is best placed to design sweeping new global institutions, particularly one requiring deep US-China cooperation, which many view as unrealistic given current geopolitical tensions, and would prefer market adaptation and targeted, evidence-based intervention over speculative, large-scale intervention now.
Full account
Bill Gates has published a lengthy essay setting out a markedly bleaker view of artificial intelligence than the one he offered in similar writings back in 2023. Titled "The turbulent AI era is here. The choices we make now are critical," the roughly 6,000-word piece marks a return to public commentary after a period of relative quiet from the Microsoft co-founder on the subject, and argues that the world is badly underprepared for the scale and speed of disruption AI is set to bring. Gates writes that the shift now under way could prove to be "one of the most turbulent times in human history," capable of acting as either a great equaliser or a source of deep injustice, depending on the choices made in the near term.
A central theme of the essay is the threat AI poses to employment. Gates predicts that large numbers of jobs, across both white-collar and blue-collar sectors, will be lost permanently rather than simply reshaped, likening the coming change to a compressed, decade-long version of the shift from manual labour to office work seen over previous generations. To address this, he puts forward specific policy ideas rather than vague appeals: taxing AI systems and robots to discourage firms from replacing staff wholesale, using the proceeds to strengthen social safety nets, and setting aside a category of "human reserved" occupations that would remain off-limits to automation. He urges governments to act before joblessness spreads and public trust collapses.
Alongside the economic warning, Gates raises concerns about security risks, including the possibility that AI could be used to help design bioweapons, chemical threats or cyber-attacks, and that such capabilities could empower both non-state actors and existing powers alike. He points to the danger of autonomous weapons systems reducing human oversight of lethal force, and to the potential for AI to make the manipulation of public opinion cheaper and more effective. One report on the essay adds that Gates believes technology executives have been reluctant to be candid about these dangers, and notes that he is hoping to discuss ways of managing AI risk with China's president, Xi Jinping, later this year.
On governance, Gates argues that existing national and international institutions are ill-equipped to deal with a technology that cuts across taxation, labour, healthcare, defence and education simultaneously. He calls for governments to create dedicated bodies capable of coordinating AI policy across departments, and for a new international structure to manage risks at a global level, while acknowledging that such solutions offer no quick fix. Despite the essay's largely sombre tone, Gates maintains that it is not too late to steer AI so that its benefits outweigh its harms, though commentators note that the piece is stronger on diagnosis than on genuinely new remedies.
Where outlets differ
One outlet frames the essay primarily as a set of economic and governance 'takeaways', giving most weight to Gates's concrete labour-market proposals (an AI/robot tax and 'Human Reserved' jobs) and his call for new international institutions.
The other outlet leads with the security and bioterrorism angle, foregrounding quotes about bioweapons, autonomous weapons and manipulation of public opinion, and adds claims — not present in the first account — that tech industry leaders are deliberately downplaying these risks and that Gates hopes to raise the issue with Xi Jinping in November.
The first source situates the essay against Gates's more optimistic 2023 writing on AI, describing it as a marked change in tone, while the second focuses less on that contrast and more on the specific threats described.
Minor factual details, such as Gates's age (reported as 70) and the timing of the essay's release, appear only in one of the two accounts.