Bridgewater co-CIO Greg Jensen wants companies that control more than 5% of global or US AI compute resources to face oversight modeled on the rules applied to too-big-to-fail banks. He draws the parallel to the concentration regulators failed to rein in among financial firms before 2008.
Greg Jensen, co-chief investment officer of Bridgewater Associates, wants regulators to treat dominant AI compute providers the way they treat systemically important financial institutions, or SIFIs. He is calling for oversight mechanisms modeled on the SIFI framework for any company that controls more than 5% of global or US AI compute resources.
Why the 5% threshold matters
The SIFI framework emerged after the 2008 financial crisis, when regulators recognized that some institutions had grown so large and interconnected that their failure would threaten the whole system. That designation triggers heightened capital requirements, stress testing, and regulatory scrutiny beyond what ordinary firms face.
Jensen made his remarks reported around September 17-18, 2026, following an interview with The Information. He pointed to projections that OpenAI and Anthropic together could command between 35% and 50% of the world's AI compute capacity within roughly two years. He compares the moment to February 2020, the quiet weeks before COVID-19 became a global emergency, arguing reactive regulation after an incident will cost more than proactive rules now.
Bridgewater's own AI footprint
Jensen's push carries a layer of complexity given Bridgewater's own position in AI. The firm established AIA Labs in 2023, a unit that now employs more than 80 staff and manages approximately $4.5 billion in AI-driven strategies. Bridgewater maintains that AIA Labs operates with deliberate human oversight built into its processes, a design choice that echoes Jensen's broader argument for accountability structures around influential AI systems.
What ownership caps would mean in practice
Translating a 5% threshold into policy is not simple. The SIFI designation for banks rests on a well-established global regulatory architecture, with the Financial Stability Board coordinating international standards. AI compute has no equivalent structure yet, and defining what counts as compute is itself hard: raw chip capacity, data center square footage, training runs, and inference throughput would each produce a different map of who controls what.
OpenAI and Anthropic, the companies most directly in Jensen's view, are both privately held and heavily backed by major technology and venture capital firms. Regulatory concentration limits could complicate their growth plans at a moment when both are making substantial infrastructure commitments.
Source: Crypto Briefing
Trading involves risk.
Risk Disclaimer
