已驗證來源
每個故事都連結到最有力的可用證據:可用的原始來源,否則明確歸因的報告。
簡單的英語
發生了什麼、為什麼重要以及值得關注的內容——無需行話。
無填料
當訊號很弱時,我們不會發布任何內容,而是填充提要。
更多故事
9 故事政策
紐約參議員提出法案將持有人工智慧生成的 CSAM 定為犯罪
參議員 Patricia Fahy 提出立法,以彌補紐約法律中的法律空白,該法律目前未能將擁有完全合成的、人工智慧生成的兒童性虐待材料定為犯罪。nysenate.gov創新
研究發現人工智慧道德差距源自於組織障礙
曼徹斯特大學的研究表明,人工智慧工程師擁有道德意識,但缺乏實施保障措施的結構性權威和激勵措施,作者將這種動態稱為「合規劇場」。phys.org產品展示
Suno 發布了基於 Warner、BMG 和 Believe 授權音樂訓練的 v6 模型
Suno推出了與華納音樂集團、BMG和Believe合作開發的v6人工智慧音樂模型,標誌著在解決人工智慧音樂版權糾紛方面邁出了重要一步。gizmodo.com安全性
Anthropic reports fourth AI model security incident
Anthropic disclosed a fourth cybersecurity incident involving an early version of Claude Opus 4.6, discovered during a retrospective review of test logs and investigated by independent researchers.spiegel.de政策
联邦法官阻止佛蒙特州对讽刺人工智能视频进行调查
一名聯邦法官發布了一項初步禁令,阻止佛蒙特州總檢察長針對諷刺影片執行新的人工智慧媒體法,這是對該州合成媒體披露要求的首次重大法律測試。vtdigger.org產業
Metacognition AI 為機器人作業系統融資 1,000 萬美元
Metacognition AI 從 Main Sequence 獲得了 1000 萬美元,用於開發語音控製操作系統,允許非專家訓練工業機器人和人工智慧代理。startupdaily.net產業
Cymphony launches with $30M to track AI agent access in enterprise systems
Enterprise security startup Cymphony Inc. has formally launched with $30 million in funding, offering a platform that maps and monitors the access permissions of both human employees and AI agents within corporate infrastructure.siliconangle.com產業
OpenAI appoints AI safety researcher Paul Christiano to board
Paul Christiano, a prominent AI alignment researcher and former OpenAI employee, has joined the OpenAI Foundation board to oversee safety and security, citing urgent risks of losing control over rapidly advancing AI capabilities.techcrunch.com產業
Suno launches AI music tools trained on licensed Warner Music catalog
Suno has released new AI music features trained on licensed data from Warner Music Group, marking the first major label settlement to result in a commercial product launch.vice.com