Greg Jensen, co-chief investment officer at Bridgewater Associates and early financial backer of both OpenAI and Anthropic, has sounded an alarming warning about the trajectory of artificial intelligence development. According to AI Weekly, Jensen estimated a 30 to 60 percent probability that the world will experience either a fatal accident or severe financial damage caused by AI systems within the next two years.
Speaking on Bloomberg's Odd Lots podcast in mid-September, Jensen articulated a sobering view of how societies respond to emerging technological threats. His assessment reflects growing concern among influential figures in the financial world about the pace of AI deployment relative to the maturity of safety and governance frameworks.
Waiting for Catastrophe
Jensen's central thesis centers on a historical pattern: governments and institutions rarely implement meaningful safeguards until after a crisis occurs. He predicted that despite awareness of potential risks, no substantial regulatory intervention will materialize until artificial intelligence systems cause tangible harm to people.
"Until the AI starts killing people, unfortunately, history would suggest we're not going to do anything, but we are going to face that," Jensen stated, according to the report. "That's going to happen, and it'd be much better if we started [regulatory action] earlier."
This perspective carries particular weight given Jensen's position within the investment community. As someone who has placed significant capital behind some of the most influential AI companies, his outlook suggests even those deeply embedded in the industry acknowledge serious risks on the near horizon.
The Stakes of Inaction
Jensen's warning touches on a persistent tension in artificial intelligence governance. Developers and companies often argue that premature regulation could stifle innovation and beneficial applications. Conversely, safety advocates contend that waiting for incident-driven regulation means accepting preventable harm as the cost of technological progress.
The hedge fund executive's dual perspective as both investor and risk assessor positions him uniquely to observe this dynamic. Bridgewater's investment in both OpenAI and Anthropic places Jensen within circles where technical capabilities and safety considerations are routinely debated at the highest levels.
Growing Industry Concern
Jensen's statements align with increasing vocal concern from industry leaders, researchers, and policymakers about AI development trajectories. While some executives emphasize the transformative benefits of advanced AI systems, others have begun raising flags about deployment velocity exceeding our understanding of potential failure modes.
The two-year window Jensen identified represents a relatively near-term forecast, suggesting he views the acceleration of AI capabilities and their integration into critical systems as already underway. This timeline implies that decisions made in the present moment may directly influence whether preventive safeguards emerge before incidents occur.
What Comes Next
Whether policymakers will act proactively on such warnings remains an open question. Jensen's prediction essentially challenges the premise that crisis-driven regulatory responses represent an acceptable approach to artificial intelligence governance, while simultaneously expecting that such responses will, in fact, be the default outcome.
