Bill Gates warns of 'turbulent' 20-year AI transition and severe security risks

Bill Gates has warned that the world faces a turbulent 20-year adjustment period as artificial intelligence reshapes society, cautioning that the technology is powerful enough to cause "a billion deaths" if it falls into the hands of criminals or rogue actors.

Speaking on NBC’s Meet the Press, the Microsoft co-founder said the early phase of the AI era will be highly volatile before its long-term benefits—such as lower healthcare and living costs—actually arrive. In a separate 6,000-word essay on his Gates Notes blog, he described the coming transition as "one of the most turbulent times in human history" and warned that governments are currently unprepared.

His intervention follows a moment of personal reckoning. Gates revealed that he has only been genuinely stunned by a new technology three times in his life. The first was the graphical user interface in 1980, which paved the way for modern personal computers. The second was in 2022, when OpenAI showed him a demonstration of what would become ChatGPT. The third occurred this year, when he looked closely at Anthropic’s AI coding assistant, Claude Code.

For a man who spent his youth writing the software that built a tech empire, the experience of being outperformed by an AI was disorienting.

"The idea that now these things are in a meaningful sense better than I am is like: What the hell?" Gates said in a recent interview. He co-founded Microsoft with Paul Allen in 1975, writing much of the early code himself. To find himself surpassed in the discipline he knows best was the catalyst that drove him to publish his warnings.

Why Bill Gates expects a turbulent AI transition

Gates believes the transition will take about two decades. During this window, the immediate economic disruptions will arrive far faster than the broad societal benefits.

While AI could eventually make medical treatment and daily life cheaper, Gates told host Kristen Welker that the initial years will be marked by severe labour disruptions and security threats. The most acute danger, he argued, is the weaponisation of advanced models.

"Even though it's pretty hard to get to 100%, there's never been a weapon as powerful as the combination of people with ill intent using the latest AI tools," Gates said. He warned that the strongest models, if left unregulated, could empower individuals to trigger events causing up to "a billion deaths."

To prevent this, Gates is urging law enforcement and legislators to establish global guardrails and tighter rules on data use and model safety.

How Anthropic's Claude Code stunned Microsoft's co-founder

The disruption will hit certain professions first. In his essay, titled "A turbulent AI era and critical choices to make", Gates pointed to software engineering, sales, customer support, and paralegal work as the most vulnerable sectors.

However, he noted a potential silver lining for developers. Because AI-assisted coding will make software significantly cheaper to build, the overall demand for software is likely to surge. This means actual job losses in software development might be less severe than in other fields, provided human designers still lead the process.

For now, the rapid progress of tools like Claude Code—which can write, edit, and debug code directly from a command line—has rewritten the timeline for what AI can achieve. For Gates, it is a clear sign that the transition has already begun, whether the world is ready or not.