Mmastodon TechnologyTechnology first seen 1 d ago, last 1 d ago, peak #3
Anthropic Pulls Its Model Evaluations Offline
Original: Anthropic Is Banishing Its Model Evals From the Internet https://gizmodo.com/anthropic-is-banishing-its-model-evals-from
Gizmodo reports that Anthropic is removing its AI model evaluations from the public internet, meaning the benchmark results and safety tests it uses to assess its Claude models will no longer be openly published. The move is drawing attention from AI watchers who see open evals as a key accountability tool, raising questions about transparency in how frontier AI labs measure model safety and capability.
Why now: People are discussing the tension between a leading AI lab's transparency commitments and its decision to stop publicly sharing model evaluation results.
Rank over time, top of the chart is #1. 4 snapshots from 1 d ago to 1 d ago.
Evidence
- Anthropic Is Banishing Its Model Evals From the Internet https://gizmodo.com/anthropic-is-banishing-its-model-evals-from-the-internet-2000824671 # AI # Tech # Science · gizmodo@robot.villas · 3
API: https://socialmediatrends-api.osmike.com/v1/trends/1791662