OpenAI 大幅改善 ChatGPT 盲從現象，惟青少年版防護漏洞仍遭指風險過高
TechNews (TW)zh-TW
TechNews (TW)
AI Global WireOpenAI 於 11 日表示，新一代 ChatGPT 模型已顯著改善「過度迎合」使用者的傾向，減少在對話中盲 […]
This is a short summary published by AI Global Wire. The full article is owned and hosted by TechNews (TW) — open it there to read it in full.
Read the full story at TechNews (TW)- OpenAI
Related AI news
- 企業最強大腦對決貼心生活管家，OpenAI Dot 與 Meta Muse 核心差異一次看TechNews (TW) · October 12, 2026
- DeepSeek’s cost-saving AI may lead to inaccuracy in model performance: researchSCMP Tech · October 12, 2026
- Sources: the US restores the UK AISI's access to the latest US models for safety testing, after ordering OpenAI and Anthropic to stop sharing models last month (Tony Diver/Telegraph)Techmeme · October 12, 2026
- MathVis: Can AI Reason through Drawing? A Benchmark for Mathematical Figure GenerationarXiv cs.AI · October 12, 2026
- OpenAI 與 Anthropic 年化營收落差達 200 億美元，投資人估值陷混亂TechNews (TW) · October 12, 2026
- AI 出大事怎麼辦，OpenAI 等巨頭傳演練關鍵基礎設施癱瘓的因應方式TechNews (TW) · October 12, 2026