AI 导读
弱通用 AI 已达成（至少按 2020 年设定的标准）。比近期预测早了一年 ✅Loebner 奖是弱图灵测试，GPT-4.5 已达到等效水平 ✅Winograd 由 GPT-3 通过 ✅SAT 由 GPT-4 以 75% 通过 ✅Montezuma's Revenge 被 GPT-6 攻克 https://t.co/1Ui626De6X https://t.co/kxrr1pCeQD
正文
Weakly General AI achieved (at least according to the criteria set from 2020). a year before recent predictions
✅Loebner prize was a weak Turing Test, equivalent achieved by GPT-4.5
✅Winograd passed by GPT-3
✅SAT passed at 75% by GPT-4
✅Montezuma's Revenge beaten by GPT-6 https://t.co/1Ui626De6X https://t.co/kxrr1pCeQD
来源：@emollick · x.com