Technology Sector
AI is simultaneously reshaping monetary policy, corporate governance, and the semiconductor supply chain — and the money and power are moving fast.
Anthropic — the AI company valued at $965 billion and eyeing a potentially massive IPO — has appointed former Federal Reserve Chair Ben Bernanke to its Long-Term Benefit Trust, an independent governance body (a structure that advises leadership and appoints board members but holds no equity) (CNBC Tech). Bernanke, who steered the Fed through the 2008 financial crisis and pioneered quantitative easing (large-scale central bank asset purchases to stimulate growth), will focus on helping Anthropic understand how AI is reshaping the broader economy. The appointment signals that serious institutional credibility is becoming table stakes in the AI race — a Nobel laureate economist on your governance board is a pointed message to regulators, investors, and a skeptical public alike. With an IPO potentially on the horizon, stacking the trust with heavyweights like Bernanke is as much about valuation optics as genuine oversight.
Fed Chair Kevin Warsh has named venture capitalist Marc Andreessen to a new task force on Productivity and Jobs — one of five panels examining Fed operations, spanning inflation frameworks, balance sheet policy, and communications (CNBC). The appointment puts Silicon Valley's most prominent AI evangelist inside the central bank's policy review process at a moment when AI's impact on labor productivity (output per worker) is one of the most consequential and genuinely unresolved questions in macroeconomics. For the tech sector, this matters: if the Fed's new framework concludes that AI-driven productivity gains are structurally disinflationary (meaning they push prices down over time), the implications for interest rate policy — and tech valuations — could be significant. Warsh's task forces have no firm deadline, but he has signaled changes this year, so the sector should watch for signals emerging from these panels sooner rather than later.
OpenAI CEO Sam Altman told CNBC that the company's new GPT-5.6 Sol model is 54% more token-efficient (tokens are the units of text AI models process, and you pay per token) on agentic coding tasks — a direct answer to enterprise clients scrutinizing AI ROI. The rollout of Sol, Terra, and Luna was initially limited to a small partner group at the U.S. government's request, with Commerce Secretary Lutnick and Treasury Secretary Bessent involved in the approval process — a sign of how seriously Washington is treating frontier AI governance. The efficiency gain matters because it's the clearest lever OpenAI has to justify expanding contracts at a moment when rivals Meta (Muse Spark 1.1) and SpaceX's xAI (Grok 4.5) are shipping competitive models almost weekly. OpenAI, privately valued at $852 billion, is reportedly in talks to offer the U.S. government a 5% stake — a detail Altman called inaccurate in specifics but didn't flatly deny, which tells you something about how unusual this company's trajectory remains.