The AI sector is holding its breath as Anthropic prepares for an IPO that could raise $100 billion at a $2 trillion valuation, a figure that would eclipse SpaceX's record offering. This ambitious move comes on the heels of a quarter where the company reported $16.8 billion in revenue, marking a 45% increase from the previous period. Despite slowing growth compared to the prior quarter's 140% spike, Anthropic maintains a robust gross margin of approximately 85% after adjustments, signaling financial health even as the broader market shows signs of fatigue.

Anthropic's Strategic Equity Giveaway

In a maneuver designed to court public favor, Anthropic announced a Public Benefit Fund that would allocate 5% of its equity to the public good if shareholders approve. This allocation increases by another 2% for every 50% jump in company value above an undisclosed baseline, with funds directed toward nonprofits or groups like workers displaced by AI. Co-founder Jack Clark intensified the narrative at the IMF and World Bank meetings, declaring the singularity has "now begun" and citing AI performance jumps on Humanity's Last Exam from 8% to nearly 68% and FrontierMath from 2% to 94%.

OpenAI's Espionage Web and Strategic Forks

While Anthropic looks outward, OpenAI is looking inward at a widening spy scandal involving an engineer accused of bugging an executive conference room. The FBI discovered a folder labeled "Saudi Gov Docs" on his phone, alongside evidence of contacts with the Saudi royal family and plans for permanent relocation. Simultaneously, CEO Sam Altman presented staff with three potential endgames for AI safety: a global watchdog similar to the IAEA, a CERN-style merger of top labs, or a single company achieving self-improving AI dominance. Altman reportedly favors the watchdog approach but warned that any path must include a speed limit on self-improving systems.

Wall Street's AI Buildout Anxiety

The financial underpinnings of the AI boom are showing cracks, with investors dumping CoreWeave bonds after a co-founder suggested borrowing from insurers' "unlimited" capital pools. Credit default swaps priced in a greater than 60% chance of insolvency within five years, while other neocloud stocks fell over 20% in a single week. Goldman Sachs estimates that the AI infrastructure buildout itself is contributing nearly half a point to annual inflation by driving up chip and computer prices, complicating the Federal Reserve's path amid hot inflation data linked partly to energy costs from the Iran war.

Key Takeaways

  • Anthropic's $2 trillion valuation hinges on slowing but massive revenue growth and a controversial public equity giveaway.
  • OpenAI faces a widening espionage investigation tied to Saudi Arabia while debating its long-term AI safety structure.
  • Wall Street is repricing AI infrastructure risk, with CoreWeave's debt and inflationary pressures causing significant stock volatility.
  • Meta's Muse Spark Open model demonstrates local AI viability on dual Mac Studios, signaling a potential thaw in Apple-Meta relations.

The Bottom Line

Anthropic is betting that public goodwill can offset market fragility, but with OpenAI's spy saga and rising inflation, the AI gold rush is getting expensive and dangerous.