AI Safety Debate Reignites: Gebru Says No Existential Threat, OpenAI Delays Launch, Anthropic Warns of Risks

The AI industry’s debate over safety and commercialization is heating up again. One of AI’s fiercest critics, Timnit Gebru, believes AI does not pose an “existential threat,” and that doom talk is more about founders making money than saving humanity.^[1]

Meanwhile, OpenAI said it is delaying the release of its latest model over safety concerns. The company said its latest Astra model would undergo more work to meet safety standards, and issued an apology for the way it handled the hacking of an Australian government website.^[4]

Anthropic, in turn, warned of “catastrophic” AI risks in a preview of its greatly anticipated IPO filing. The disclosures reportedly mention mounting losses, leadership proposals to retain power, and how its AI development plans could “further increase the risk that our models cause harm.” Reports also say Anthropic is eyeing a $2 trillion valuation.^[5]

On the enterprise side, AI is moving from experimentation to production, and the cost conversation is changing. One view holds that when customers talk about AI costs, the conversation usually starts with token prices and ends with access to the latest, most capable model in the cloud; but enterprises do not necessarily always need that level of capability, and model choice is only one part of the equation.^[3]