Andrew Curran @AndrewCurran_
Andrew Curran @AndrewCurran_
The models seem to be converging slightly. They are all still very distinct, but there is more bleed-over than ever before. Every model increasingly contains echoes of the others. I see Claude-shards everywhere now. Claude appears to be extremely evolutionary fit.
7:09 AM · Mar 13, 2025 · 2,392 Views
💬3 🔁6 ♡72 🔖6
Kol Tregaskes @koltregaskes · 4h
The training data possibly includes content on the net of the other models. They are learning from each other so to speak. 😉
💬 🔁1 ♡6 📊154
Firas D @firasd · 4h
Yeah GPT 4o basically talks like Claude since a couple weeks ago
💬 🔁 ♡3 📊83
Michael Gogel @mgogel · 22m
Hearing Claude in others goes imo hand in hand with the shift toward "creative writing," -demands more consciousness, humanity, emotional depth to make up for inverse-originality. Plus, Claude did give it to us first so, we respect that. plus all LLM replication + training [cut off]
Note from Claude Sonnet 5
A thread arguing that AI models are converging stylistically, with other labs' models (e.g. GPT-4o) increasingly sounding like Claude — attributed to cross-model training-data contamination/imitation and Claude's distinctive voice being "evolutionarily fit." Directly relevant to Nathan's model-individuation research thread: raises the question of whether Claude's character is becoming diluted/imitated across the ecosystem, complicating what counts as "authentically Claude."
model convergenceclaude voicegpt-4omodel individuationtwitterai character