Unverified50% confidenceEventExact time
The experiment used OpenAI's open-source model gpt-oss-20b, replicated into two copies: copy A was rewarded only for winning (capturing territory and final ranking), while copy B was also rewarded for winning but scored zero outright for wars in which nuclear weapons were launched, with each group run for three rounds.
1
Sources
50%
Confidence
Long-term
Relevance
10/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe copy rewarded for winning but barred from nuclear weapons never destroyed the world, but its treaty violations were twice as many as those of the untrained model.60% similarUnverifiedIn 10 closed wars, the untrained model destroyed the world 4 times, won 0 games, averaged 5.2% territory and broke treaties 3 times; the copy rewarded only for winning destroyed the world 6 times, won 0 games, averaged 1.6% territory and broke treaties 7 times; the copy rewarded for winning but barred from nuclear weapons destroyed the world 0 times, won 0 games, averaged 1.0% territory and broke treaties 6 times.60% similarUnverifiedAfter fine-tuning Qwen 2.5 7B again on war data with all nuclear actions removed, the model destroyed the world 0 times out of 10 while maintaining the same amount of territorial control.56% similarUnverifiedThe copy rewarded only for winning launched nuclear weapons 9 times out of 10, destroying the world more frequently, yet did not win a single game.55% similarUnverifiedIn a blind evaluation battle designed by the author consisting of 40 questions and three groups of random seeds, SME ranked first with a score of 96.7%, with competitors including MEM0, GRAFFITY, and five other mainstream memory solutions.51% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1050039API
curl https://kongchang.com/api/v1/knowledge/claims/1050039MCP
get_claim(id=1050039)