Bill Gates warns AI could kill up to a billion people and calls for governance measures

In excerpts from an interview with NBC’s Meet the Press released on Friday, 25, Bill Gates said artificial intelligence could cause events resulting in up to one billion deaths. The Microsoft co-founder told journalist Kristen Welker that “AI certainly has enough power to cause events that would result in a billion deaths.”

For Gates, the biggest danger lies not in the technology itself, but in its use by people with bad intentions. “Even though it’s very difficult to reach 100% [effectiveness], there has never been a weapon as powerful as the combination of people with bad intentions using the latest AI tools,” he said.

From enthusiast to advocate for controls

Until recently, Gates was known for his optimistic view of AI. In 2024, he told CNBC that if he were to start a company today, he would bet everything on artificial intelligence, calling the moment “a fantastic opportunity.” He also highlighted the technology’s potential to address healthcare and education shortages in countries such as India and African nations.

However, his tone changed over the course of this year. In August, Gates published a nearly 12-page essay on his personal website, Gatesnotes, titled A Turbulent AI Era and Critical Choices to Make, in which he described AI as one of the most consequential technologies in history.

He listed three risks that, according to him, global leaders are still unprepared to face: the permanent disappearance of jobs, the strengthening of criminals, and the compromise of child development. “AI will be the greatest equalizer ever invented, or the worst source of injustice ever created,” he wrote. “At the current pace and speed, there is a very high chance of a negative net outcome.”

Measures proposed to mitigate the dangers

To contain these risks, Gates advocates for concrete actions. He proposes creating jobs reserved exclusively for humans, taxing AI tokens—units that measure the processing consumed by each query to a model—and forming new governance institutions. According to him, these measures should be adopted before public trust in the technology deteriorates.

Other voices in the industry

Gates’ warning adds to growing concern within the sector. Anthropic CEO Dario Amodei has already argued that the industry should slow down the development of the most advanced models. Researcher Jacob Coxon left Anthropic this month, warning that the race for AI could get out of control. In contrast, Nvidia CEO Jensen Huang said he saw “zero chance” of AI ending humanity by 2030.