AI Safety Report Card: Top Firms Fail to Protect Humanity

Imagine the architects of our future getting a failing grade on safety. A new AI Safety Index ranks top firms on their efforts to prevent catastrophic risks. Even the best performers, OpenAI and Anthropic, only scored a C+, while giants like Meta and Elon Musk’s xAI received a dismal D. Crucially, every single company failed on existential safety, showing no credible plans to prevent catastrophic misuse or loss of control of their powerful systems. This highlights a dangerous, unregulated "race to the bottom" where innovation is dangerously outpacing crucial safety protocols, risking everything from bioweapon creation to government destabilization. This index serves as a critical tool for policymakers to push for binding safety standards and for the public to demand greater accountability from tech giants.