Artificial Analysisがベンチマークテスト更新 非公開データ拡充で「ゲーム化を防ぐ」
ITmedia AI+ja
Artificial Analysisが知能指標「Intelligence Index」をv4.2に更新。実務に近い評価を加え、非公開データの比重を倍増して「ゲーム化を防ぐ」とうたう。首位はClaude Fable 5.1で、GPT-6 Astraが続く。
This is a short summary published by AI Global Wire. The full article is owned and hosted by ITmedia AI+ — open it there to read it in full.
Read the full story at ITmedia AI+- OpenAI
- Anthropic
Related AI news
- Anthropic IPO set to put AI governance under public-market scrutinyDIGITIMES · September 7, 2026
- 擅將報導文章用於訓練 AI,OpenAI 微軟再遭美兩報社提告TechNews (TW) · September 7, 2026
- 擅將報導文章用於訓練 AI,OpenAI 微軟再遭美 2 報社提告TechNews (TW) · September 7, 2026
- Seattle Times and Newsday sue OpenAI and Microsoft for infringementThe Verge AI · September 6, 2026
- SaaS 巨頭 Salesforce 震撼攜手 Anthropic,為何甘願當「笨水管」把用戶入口讓給 Claude?TechNews (TW) · September 6, 2026
- Authors push back as publishers and agents make claims on Anthropic settlementTechCrunch AI · September 6, 2026