Norm 👨‍🦼

ANTHROPIC WANTS BILLIONS FOR AI IT WARNS COULD THREATEN HUMANITY Anthropic, the company behind Claude, has filed to go public in what could become the largest IPO in history. Its pitch includes an extraordinary warning: advanced AI could pose catastrophic, even existential, risks to humanity. The company reported $16.3 billion in revenue in the first half of 2026, fourteen times the year-earlier figure, along with an $80.7 billion net loss. Most of that loss came from a one-time accounting charge; its operating loss was roughly $693 million. Reports put its commitments for chips, data centers, and computing power at up to $807 billion over the next decade, though some deals remain unfinished. Why pour so much into making AI more powerful when its builders warn they may lose control? CEO Dario Amodei, who has estimated a 10–25% chance of AI seriously derailing human civilization, says the goal is to reduce those risks. Anthropic warns about biological weapons, cyberattacks, authoritarian surveillance, economic disruption, and advanced systems escaping human control, even as its AI helps develop more advanced versions of itself. These disclosures do not mean catastrophe is imminent. AI could also accelerate medical breakthroughs and scientific research, and SEC filings must disclose potential risks. Anthropic says its governance structure protects that safety mission: its seven cofounders would retain majority voting power over certain decisions, while a special trust would have authority to elect most of the board. But that protection also concentrates enormous influence in relatively few hands. Who decides how far this technology should go, and who holds those decision-makers accountable? Anthropic is asking investors to finance increasingly powerful AI while warning about the consequences if control fails. The question is not just whether the investment pays off. If something goes wrong, will the decision-makers bear the cost, or will everyone else? Larry Ward joins me to break this down.