SKEPTICISM GROWS OVER OPENAI'S AI DANGER CLAIMS

OpenAI's pattern of announcing AI safety concerns without releasing supporting evidence deserves scrutiny, according to analysis by technologist John Thickstun. The company declared GPT-2 too risky to release in 2019, citing abuse potential. Yet the model eventually became public with minimal incident, raising questions about the original threat assessment. Thickstun argues that OpenAI benefits from portraying its systems as inherently dangerous. When the company emphasizes AI risks, investors hear about power and capability—potentially driving valuations higher. This creates misaligned incentives between safety messaging and financial interests. The dynamic matters as OpenAI and competitors face increasing scrutiny over AI safety claims. Public announcements about "rogue agents" or capability breakthroughs shape market perception and regulatory discussions. The criticism suggests tech companies should separate safety research from marketing narratives, ensuring danger claims receive independent verification rather than serving dual purposes.