Anthropic has issued a rare warning in its IPO prospectus that its AI technology could pose catastrophic or existential risks to humanity. Nearly one-third of the prospectus is dedicated to elaborating on risk factors, revealing that the company allocates only about 6% of its computational resources to safety research. Meanwhile, it cites researchers' views that the probability of AI causing human deaths within a decade exceeds 10%. AI safety governance faces numerous challenges, such as immediate costs with unpredictable benefits and effects, ineffective safety assessments, conflicts between development speed and safety investments, and frequent industry-related incidents. Anthropic believes that building safe AI is a shared responsibility of society and that the market will provide corresponding rewards. However, some investors argue that Anthropic's emphasis on safety is essentially aimed at serving the growth of its B2B business. Recently, Anthropic's valuation and annualized revenue have grown rapidly, with its customer base gradually shifting toward enterprises and its product offerings continuously expanding. Given its leading model capabilities, safety is becoming a key commercial selling point to attract enterprise clients.