Anthropic Opposes the Extreme AI Liability Bill That OpenAI Backed
Claude maker fights bill that would protect labs from billion-dollar disaster liability.
Anthropic has publicly opposed Illinois Senate Bill 3444, legislation backed by rival OpenAI that would provide AI developers with significant liability protection. The bill would shield labs from legal responsibility if their systems are used to cause catastrophic harm—defined as mass casualties or over $1 billion in property damage—provided the company has published a safety framework. Anthropic's head of US state and local government relations, Cesar Fernandez, called the proposal a "get-out-of-jail-free card" that undermines public safety and accountability.
Behind the scenes, Anthropic has been lobbying Illinois state senator Bill Cunningham, the bill's sponsor, to either substantially amend or kill the legislation. The company testified in favor of a separate, stronger Illinois bill (SB 3261) that would require frontier AI developers to create public safety and child protection plans. Meanwhile, OpenAI argues SB 3444 reduces risks while allowing technology adoption, stating it's working toward a "harmonized" state-level approach in the absence of federal action. Policy experts warn the bill would dismantle existing common law liability that incentivizes companies to mitigate foreseeable risks.
- SB 3444 would protect AI labs from liability for catastrophic misuse causing mass casualties or over $1B in damages.
- Anthropic is lobbying against the bill, calling for "real accountability," while OpenAI supports it as a balanced framework.
- The fight exposes a major regulatory split between the two leading AI labs as state-level lobbying intensifies.
Why It Matters
This clash sets the precedent for whether AI companies bear legal responsibility when their powerful models are weaponized.