Bill Gates Warns AI Could Transform Jobs and Outpace Human Intelligence

Artificial intelligence could reshape employment, scientific research and economic production over the next two decades, Microsoft co-founder Bill Gates warned in interviews published in October 2026. Gates said increasingly capable AI systems could surpass human performance in several fields, intensify cybersecurity threats and reduce demand for workers, raising questions about how governments should manage the technology's expansion.

In an interview published by TokenPost on October 22, Gates estimated that a broad economic transition driven by AI could take approximately 20 years. He described that estimate as "a reasonably safe number," while acknowledging that the transformation could happen faster or take longer.

Gates distinguished this economy-wide shift from changes expected within individual industries. In an essay published August 26, he said AI could reshape law, customer service, medicine, software and manufacturing over the following decade.

He also warned that job-market disruption and cyberattacks could produce significant negative effects within two to three years of the interview. The forecast covered a relative timeframe rather than specifying calendar years for those developments.

During an interview with The Observer published October 25, Gates described AI as an evolutionary turning point that could eventually move intelligence beyond the limits of human biology.

"The most intelligent thing is being redefined," Gates said.

Asked whether that meant humans would no longer represent the highest form of intelligence, he answered directly.

"Right," Gates said.

Gates said AI could outperform people in mathematical breakthroughs, biological simulations and engineering design within the next 20 years. He also pointed to recent AI-assisted mathematical research that addressed 372 problems that had challenged mathematics professors for years.

He said his own experience with software development had already demonstrated the pace of change.

"Late last year was when my own personal coding abilities, which I obsessed on starting at age 13 – I could see that those were essentially obsolete," Gates said.

Gates predicted that AI and robotics could eventually automate intellectual work and physical tasks. He described robots building houses, growing food and performing surgery as possible future applications, not capabilities that have already been universally achieved.

"Skills of invention, skills of white-collar work and then, with humanoids, the skills of physical human labour," Gates said, describing the areas AI could eventually replace.

He also questioned how people might find purpose if machines could meet many economic needs that have traditionally depended on human labor.

"It's something that, you know, philosophers, religious people, the policy people, really need to be engaged in," Gates said.

As one possible reference point, Gates cited communities that deliberately limit their use of advanced technology.

"There are groups of humans like the Amish who don't choose to live at the technological frontier," Gates said.

Alongside his economic forecasts, Gates has called for stronger safeguards governing AI development. In an interview with The New York Times columnist Ezra Klein in September, he compared the potential risks to dangers that society manages through safety rules in medicine, aviation and transportation.

"This is the most dangerous thing that humans have ever gone near," Gates told The New York Times.

He argued that companies should not be allowed to introduce powerful systems without adequate protections and questioned whether lawsuits after harm occurs could substitute for preventive regulation.

"In other areas, do we just say, Hey, release your drugs? There's no FDA, there's no airline safety board, there's no requirement that cars use seatbelts. Do we just use the liability laws to try and keep humans safe?

Oh, you're shipping opioids. Somebody should just sue you," Gates said.

Gates characterized reliance on voluntary industry oversight without sufficient safeguards as "just insane." His concerns came amid a broader debate over whether governments should establish mandatory standards for increasingly capable AI systems.

A Quinnipiac University poll of 1,202 US adults conducted September 24-27 found that 91% considered federal AI safeguards very or somewhat important. The survey also found that 81% believed safety should take priority over innovation, while 53% expected AI to cause more harm than good in their daily lives.

The poll was released one day after President Donald Trump announced an agreement with technology executives centered on voluntary industry measures, including third-party safety assessments and stronger internal controls. The agreement did not introduce new federal rules.

Public concerns also extended to the infrastructure required to support AI. The Quinnipiac survey found that 72% of Americans opposed a new data center near their homes, up from 65% in a March poll, as the technology's expansion increased demand for reliable electricity and additional power infrastructure.

Gates also warned that AI could increase the scale and sophistication of cyberattacks. His earlier concerns included extortion and other malicious activity, while his broader forecasts described a future in which AI could make many goods and services more abundant through automation.