不過(guò) ,現(xiàn)在最讓AI閱讀器頭疼的縫隙是所謂的“提示注入進(jìn)犯”(prompt injection attacks)。微軟也將分別在自家的Chrome 、AI閱讀器究竟該何去何從呢?
參閱鏈接
[1]https://x.com/AnthropicAI/status/1960417002469908903
[2]https://claude.ai/chrome
[3]https://www.anthropic.com/news/claude-for-chrome
[4]https://techcrunch.com/2025/08/26/anthropic-launches-a-claude-ai-agent-that-lives-in-chrome/
本文來(lái)自微信大眾號(hào)“量子位”,Claude經(jīng)過(guò)小規(guī)模試點(diǎn)搜集數(shù)據(jù)和安全反應(yīng) ,
還能幫你回郵件