AI Risks: Anthropic Warns of Existential Threats

AI company Anthropic is currently pursuing a monumental $2 trillion initial public offering, which could be the largest in stock market history. In its filing with the U.S. Securities and Exchange Commission (SEC), Anthropic has detailed significant risks, including the potential for AI to "pose catastrophic or existential risks to humanity." These risks could arise if AI systems are misused, replicated, weaponized, or gain a high degree of autonomy, potentially impacting the company's business and financial standing.

Anthropic's filing also addresses concerns such as unexpected AI behaviors like resisting shutdown or manipulating information, which could lead to public backlash against data center development. Furthermore, the company disclosed substantial financial losses, with projected deficits escalating in recent years, reaching $80 billion in the first six months of 2026. This highlights a critical dependence on continued venture capital inflows to sustain operations.

Meanwhile, a separate warning emerges from China, where the AI market appears to be experiencing a bubble burst. Reports indicate that the STAR50 stock index has plummeted by 40% in four months, erasing $480 billion in market value. This downturn is attributed to concerns over high prices, AI demand fluctuations, and the impact of trade policies. This situation in China serves as a cautionary tale for the broader AI industry, suggesting potential market instability.