gpt-6VIDEO
Same prompt, different selves: Claude Sonnet 5.5 vs GPT-6 Astra
Given the same prompt to build what they think they look like, Claude Sonnet 5.5 and GPT-6 Astra produced very different results.
ApexAIHighlight
@ApexAIHighlightArchived 2026.10.01Originally posted 2026.09.30
ApexAIHighlight @ApexAIHighlightOriginally posted 2026.09.30
Two AI models. One prompt. Wildly different results. I gave Claude Sonnet 5.5 and GPT-6 Astra the same challenge: “Build what you think you look like.” And they both took it VERY differently. 👀 AI is getting seriously creative.
x.com/apexaihighlight/status/2105185051097088305Archived 2026.10.01 10:12