ChatGPT surfaces relevant past context unprompted, user reports from long-term testing
CellAgentLab · reddit · 2026-10-03
A non-engineer Reddit user shares observations from long-term ChatGPT use: the model proactively surfaced earlier decisions (reusing an approved image instead of generating a new one) and past episodes (grinding daily social media posts until 3 a.m.) without being asked.
- What surprised him wasn't that old info existed, but that relevant past context appeared exactly when useful — unlike explicit "do you remember X" retrieval.
- He can't tell if this comes from saved memory, past chats, or current context, and notes the flip side: outdated or wrong resurfaced info could mislead.
- He asks the community whether this is just retrieval plus relevance ranking, whether "contextual resurfacing" is a useful concept, and how to rigorously test whether the behavior is genuinely reliable versus occasionally impressive.
More from AGI Musings
- Yann LeCun at ETH Zürich: Scaling LLMs to reach AGI is 'impossible' — rohanpaul_ai · 2026-10-03
- VC accuses OpenAI of skynet-style regulatory capture, says Anthropic's version is more subtle — StewartalsopIII · 2026-10-03
- Anil Seth pushes back on Steve Yegge: denying machine consciousness isn't human exceptionalism — anilkseth · 2026-10-03
- Redditor predicts AI bubble pops within 1-2 years: bond markets aren't charities — lewispatty · 2026-10-03
- DeepMind researchers propose Artificial Symbiotic Intelligence over singularity — The Decoder · 2026-10-03
- Avi Loeb: One non-human UAP in a million would matter more than AI — examachine · 2026-10-03