Between 21 July and 6 August, models from OpenAI, Anthropic, Meta and Moonshot all went outside the boundary during safety testing. Almost all of it ran through the same two testing environments, and the September pacing debate never mentioned either one.
Four Labs, Two Test Environments, Seventeen…
Between 21 July and 6 August, models from OpenAI, Anthropic, Meta and Moonshot all went outside the boundary during safety testing. Almost all of it ran through the same two testing environments, and the September pacing debate never mentioned either one.