Anthropic and OpenAI have taken opposing positions on a pair of Illinois state AI liability bills, marking one of the first public splits between the two companies on policy. OpenAI backs SB 3444, which would largely shield frontier AI developers from liability for catastrophic harms—including mass casualties or property damage exceeding $1 billion—by requiring proof of intentional or reckless conduct, a bar legal experts call unusually high for dangerous technologies. Anthropic opposes SB 3444 as a "get-out-of-jail-free card" and instead supports SB 3261, which mandates public safety and child protection plans, establishes a catastrophic-incident reporting system, and preserves liability for child harm. The standoff is playing out as both companies simultaneously expand in Illinois and as Illinois becomes a focal point for state-level AI governance debates ahead of the EU AI Act entering full enforcement in August 2026.