Primary sourcelmarena.ai
View sourceEvent ·Multi-source
AI模型更倾向于选择自己的答案
研究人员比较了12个模型在1,460场对决中的34,580次判断结果。AI模型平均选择自己答案的频率为58%,而人类仅为34%。GPT-6 Astra选择自己的比例高达88%。AI模型之间的一致性达到79%,而与人类的一致性仅为57%。GPT-5.6 Sol在96%的情况下都选择胜者,而人类有32%的情况选择平局或两者都不好。
Current read
Arena研究发现AI模型有自己独特的判断偏好,与人类明显不同,GPT-6 Astra甚至88%选择自己。
11 sources73° AI heatLast updated 2026/9/29 16:19:03
Evidence
11 sourcesRelated source 1Decoder
View sourceRelated source 2IT之家
View sourceRelated source 3arXiv cs.AI
View sourceRelated source 4Perplexity
View sourceRelated source 5kimmonismus
View sourceRelated source 6Andrew Curran
View sourceRelated source 7Lxfater
View sourceRelated source 8OpenAI Blog
View sourceRelated source 9The Rundown AI
View sourceRelated source 10theverge
View sourceConflict check
Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.