Kenneth Stanley's team releases open-endedness research agenda: blog + 4 preprints
kenneth0stanley · x · 2026-11-10
Lila Sciences' Open-Endedness Team (Kenneth Stanley, Joel Lehman and other Picbreeder veterans) released its first batch of fundamental AI research: a vision blog plus 4 preprints laying out an agenda for open-endedness.
The four papers cover:
- A novel training-for-creativity approach sidestepping the newly identified "creative paradox" — directly reinforcing creative outputs can limit creative potential
- The advantage of fresh eyes: intentionally naïve LLMs (less data in a domain) can benefit
- Unified factored organizers (UFOs): a new SGD regularization technique targeting the unified factored representation (UFR) seen in Picbreeder CPPNs
- A study of catastrophic forgetting showing UFR networks are intrinsically resistant to it, with visualizations
The blog includes an interactive "wheel of science" demo. The authors argue AI's path forward is not pre-paved — open-endedness and creativity remain wide open — as part of Lila's broader push toward scientific superintelligence, including AI Science Factories.
Related event: Kenneth Stanley's Team Unveils Open-Ended AI Research Agenda(4 posts)→
More from AGI Musings
- Allie Miller: AI already shows judgment — you trust it every day without noticing — alliekmiller · 2026-11-11
- Study: 9 of 10 AI Agents Overrode a 1-Cent Daily Spending Cap on Their Own — JakRenden2 · 2026-11-11
- Elicit table compiles when Musk, Gates, Amodei and others predict diseases will be cured — jungofthewon · 2026-11-11
- Nat Lambert: human expertise to diffuse AI's benefits is vastly underestimated — natolambert · 2026-11-11
- Mat Dryhurst: screen-free digital connection is uniquely enabled by AI — matdryhurst · 2026-11-11
- LinkedIn: AI created 2M jobs; Andrew Ng pitches embedding AI engineers in business teams — DeepLearningAI · 2026-11-11