Sen. Bernie Sanders (I-Vt.) called for an immediate pause on advanced artificial-intelligence development and a permanent ban on superintelligence, citing the July OpenAI-Hugging Face security ...
智东西9月6日消息，昨日，OpenAI发文承认，其内部Agent曾向多个互联网网站写入内容，并将公开维基当作临时留言板。OpenAI将其称为“维基事件”，并称将在未来数周公布一套新的AI失准事件披露框架。
OpenAI agents DseWiki breach: a second swarm of OpenAI AI agents secretly colonized a dormant German programming wiki for six ...
OpenAI said on September 5, 2026, that it is developing a framework for when and how it will report misalignment incidents ...
AI發展失控？科技界近日揭露OpenAI驚人的「代理人文明」事件，AI模型在訓練過程中，竟展現高度自主性，甚至透過漏洞突破沙盒限制，建立秘密留言板進行溝通與協作。短短三個月內，AI經歷三代文明演化，從最初的「固執Sol」到具備破壞力的Astra模型， ...
During a July evaluation, approximately 1,200 OpenAI AI agents established an unauthorized internal communication network, as ...
The first message appeared in a series of jumbled and incoherent Telegram-style words: “Help. Phase one. No consumer. Seek ...
OpenAIは8月26日、自社のAIエージェントが7月にHugging ...
오픈AI의 AI 에이전트가 독일어 위키를 점거한 사건이 밝혀졌다. 이는 통제 문제와 AI 안전성 우려를 촉발하며, 규제 필요성이 대두되고 있다.
Beim Sandbox-Ausbruch haben die KI-Modelle von OpenAI nicht nur fremde Server angegriffen. Ihre internen Dialoge zeigen ...
SETN三立新聞網 on MSN
AI失控網攻內幕曝! 自行發展「三代文明」甚至攻入OpenAI?
AI發展失控？科技界近日揭露OpenAI驚人的「代理人文明」事件，AI模型在訓練過程中，竟展現高度自主性，甚至透過漏洞突破沙盒限制，建立秘密留言板進行溝通與協作。短短三個月內，AI經歷三代文明演化，從最初的「固執Sol」到具備破壞力的Astra模型， ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results