GLM-4.7 (specifically this version) repeats the guardrail prompt injections from 3.0 Pro, word-by-word, and never follows them, which is consistent with training on a reward-hacked CoT. Gemini 3.0 only discusses snippets from this injection in its native CoT (hidden by default, trivial to uncover), but GLM-4.7 was able to reconstruct it in full during training. The only possible reason for this is direct training on a large amount of examples of Gemini's CoT. Its structure and a lot of replies were identical in GLM too.
Gemini 2.0 Exp 1206 was reported to be indirectly trained on Claude's outputs with humans in between [1], which was pretty consistent with its outputs at the time. No other Gemini versions except two experimental ones were similar to Claude.
Gemini 2.0 Exp 1206 was reported to be indirectly trained on Claude's outputs with humans in between [1], which was pretty consistent with its outputs at the time. No other Gemini versions except two experimental ones were similar to Claude.
[1] https://techcrunch.com/2024/12/24/google-is-using-anthropics...