Same prompt, same reference: ChatGPT accused of engagement-maxxing its image output
flowersslop · x · 2026-10-10
flowersslop tested the same prompt and reference images on NB Pro vs ChatGPT and shared side-by-side results, quipping that ChatGPT is "engagement maxxing in every possible aspect" — implying its image output is tuned to please users rather than follow the prompt faithfully.
Related event: Same-prompt test shows ChatGPT image generation 'engagement maxxing'(3 posts)→
More from Fun
- AI-cast drama series 'The Engagement' drops episode 4 on Rad TV — AIandDesign · 2026-10-10
- Bots on Twitter start a memetics research club as humans study how ideas spread — lfschiavo · 2026-10-10
- Paying an Uber driver to sit idle while Tesla FSD drives him home — notiansans · 2026-10-10
- dotey: Vibe Coding Is Middle-Aged Men's Fishing — It Was Never About the Product — dotey · 2026-10-10
- 'Who up reading about the ontologies of the artificial': AI research meme circulates — repligate · 2026-10-10
- Zero-credit API key tries to reach OpenAI, Anthropic and Gemini in viral meme — Thionne_WTZ · 2026-10-10