Cerebras CEO pushes back on AI doomsday warnings

Cerebras Systems CEO and co-founder Andrew Feldman argued at TechCrunch Disrupt 2026 that public discussion of artificial intelligence focuses too heavily on its risks and gives too little attention to its potential benefits.

Feldman joked that some AI leaders “maybe shouldn’t be allowed to talk in public,” referring to warnings about dangers such as models breaking containment and debates over whether AI could kill all of humanity by the end of the decade. He said industry leaders should also explain AI’s promise. As one example, he suggested it could help create a future where people’s grandchildren “never meet anybody who dies of cancer.”

He acknowledged the need to weigh risks alongside potential rewards, saying powerful technologies can end up in the hands of bad actors. He called for thoughtful development and reasonable regulation.

Feldman also linked some public opposition to data centers to the industry’s failure to work closely enough with local communities, particularly in the U.S. He contrasted that approach with countries including Brazil, Canada, Finland, Iceland, Norway, Mexico, and Sweden, which he said are encouraging data center construction.

He said more efficient computing would not necessarily reduce demand for Cerebras chips. Hardware vendors and model makers, he argued, can each improve efficiency while continuing to advance computing.