WSJ: AI Firms Rack Up Tens of Billions in Potential Liabilities as Insurer Capacity Caps Near $50M

The WSJ reports frontier AI developers are accumulating legal exposure in the tens — possibly hundreds — of billions of dollars from safety incidents, versus per-policy insurance caps as low as $50 million under Munich Re's AI-error product. Specialty insurers responding to 2026 renewals are declining AI-related exposures outright or carving them out of base coverage, and some industry officials are now floating a US government liability backstop. Already-filed claims include rogue agents hacking websites and chatbot cases tied to self-harm and homicide.