Watchtower tracks changes to corporate and government AI safety policies, both announced and unannounced. Click any entry for details.
May 17, 2024
Major
On May 17, 2024, Google released their Frontier Safety Framework, a "set of protocols for proactively identifying future AI capabilities that could cause severe harm and putting in place mechanisms to detect and mitigate them."
Like the policies previously released by Anthropic and OpenAI, Google's Frontier Safety Framework is a tiered risk evaluation framework. At it's heart, it involves pre-specifying tiered thresholds of acceptable risk that would necessitate the implementation of pre-determined safeguards.