AI's 'Existential Race' Rings Alarm: Former OpenAI Researcher Warns of Uncontrolled Future

Daniel Kokotajlo, a former governance researcher at OpenAI and current executive director of the AI Futures Project, has issued a stark warning: AI companies are locked in a dangerous 'existential race' with each other, prioritizing speed over safety and creating systems that cannot be fully trusted. His concerns stem from the rapid, unchecked development of advanced AI, which he believes could lead to a future where humans lose control. This relentless competition, driven by market share and the push to develop Artificial General Intelligence (AGI) and even Superintelligence, sees tech giants like OpenAI and Anthropic constantly trying to outdo each other, building 'moats' to keep users engaged. Kokotajlo, known for his dire 'AI 2027' scenario predicting potential human extinction from AI, is now pushing 'AI 2040: Plan A', a policy roadmap for international transparency and slowing down AI development. As the industry hurtles forward, the challenge for global policymakers to create effective regulation grows more urgent. The EU AI Act is gradually coming into force, and other regions like the US and China are trying to establish their own rules, but the regulatory landscape remains fragmented and struggles to keep pace. The core debate now revolves around whether safeguards can be established before the 'existential race' leads to irreversible consequences, or if the pursuit of profit will ultimately win over caution.