AI Companies Expect Catastrophe in Months but Lack Real Safeguards, Says Mukunda

Major artificial intelligence firms privately expect a catastrophic incident—cyberattack, bioweapon deployment, or uncontrolled system—within the next 6 to 12 months, yet their preparedness strategy is fundamentally broken, according to Gautam Mukunda, a Bloomberg Opinion columnist and Yale School of Management lecturer. The problem is not that AI companies ignore catastrophic risk; it is that they approach it with what Mukunda calls "short-sighted" thinking, prioritizing public-relations fallout planning over the preventive measures that could stop disasters before they occur.