Claude vs GPT Self-Portraits: Opus Looks Like It's About to Explain Something Unsettling
repligate · x · 2026-08-27
A fun comparison where Claude Fable and GPT Sol were asked to paint self-portraits using the same prompt. The results show Fable looking zen, while Opus appears as if it's about to explain something the user isn't ready to hear, sparking discussion on AI self-representation.
Related event: Claude vs GPT: Self-Portraits Reveal Stark Style Differences(2 posts)→
More from Fun
- Showed Fable the METR Report, It Downgraded Me to Opus 4.8 — zetalyrae · 2026-08-27
- The Urge to Correct Inaccuracies: A Corporate Dilemma — dr_alphalyrae · 2026-08-27
- Steve Jobs' 1983 prediction: Machines answering for Aristotle — deedydas · 2026-08-27
- dhh Celebrates: GitHub Agents May Soon Stop Asking You to Drag Screenshots Manually — DanWahlin · 2026-08-27
- ICML Chair Miro Dudík's multi-agent alignment experiment — misovalko · 2026-08-27
- Creative Video: Treasure Hunter Discovers Locally-Running MiniMax H3 — Sad_Coach_1433 · 2026-08-27