Watchtower tracks changes to corporate and government AI safety policies, both announced and unannounced. Click any entry for details.
OpenAI
Apr 15, 2025
Major
Updated its preparedness framework
Anthropic
Mar 31, 2025
Moderate
Released v2.1 of their Responsible Scaling Policy
Mar 6, 2025
Slight
Unannounced
Scrubbed mentions of diversity and equity from the mission description of their Responsible AI team.
Anthropic
Feb 27, 2025
Slight
Unannounced
Removed "White House's Voluntary Commitments for Safe, Secure, and Trustworthy AI"
Feb 4, 2025
Major
Removed previous commitment not to develop AI for use in warfare or surveillance
Feb 4, 2025
Major
On February 4th, Google released a new version of their Frontier Safety Framework.
Meta
Feb 3, 2025
Major
Released a responsible scaling policy, entitled their Frontier AI Framework.
OpenAI
Jan 14, 2025
Slight
Unannounced
Adjusted the language on the o1 system card webpage, changing "o1" to "o1-preview."