#大佬观点
#大佬观点
今日 38 条
-Zho-@ZHO_ZHO_ZHOAI 评分1313
Rohan Paul@rohanpaul_aiAI 评分2323
Rohan Paul@rohanpaul_aiAI 评分1313如果 OpenAI 的 CFO 在 CNBC 上说了“Muse”,我正在想象他们早上的站会。😅

Ethan Mollick@emollickAI 评分3232
jason@jxnlcoAI 评分1414
引用Alexandr Wang@alexandr_wangcredit where credit is due: they really dodged the butthole allegations
Elon Musk@elonmuskAI 评分1818
dex@dexhorthyAI 评分1313听说你们喜欢垃圾内容(另外是的,Opus 5.5 确实不错,我知道我知道)

Alexandr Wang@alexandr_wangAI 评分88
jason@jxnlcoAI 评分1010
Elon Musk@elonmuskAI 评分2222引用jimmah@jamesdoumaThis really brought home to me this new machine creativity. I don’t know the recipe, but it’s more than vibes I see. There’s substance in the lyrics and there’s nuance for the critics And a story being told in this AI homily. And a message being sold by the Ai homily.
Tibo@thsottiauxAI 评分3232
Tibo@thsottiauxAI 评分88
Diogo Almeida@CompleteSkepticAI 评分1313
Replit ⠕@ReplitAI 评分3434
elvis@omarsar0AI 评分2626Gary Marcus:The Road to AI We Can Trust(RSS)AI 评分5959 Gary Marcus 批评白宫“超级智能”协议缺乏实质约束
Gary Marcus 评析特朗普政府发布的白宫“超级智能”协议,称其承诺的四层控制与审计本质上是“不受监管、不给公众发声”的自我监管。他质疑协议中“独立”审计人的含义,并指出两周前业界谈论的 AI 放缓(Pacing)议题未体现在协议中,称 Dario、Sam 和 Elon 都退缩了。
Alexandr Wang@alexandr_wangAI 评分1515
François Chollet@fcholletAI 评分2727
Nathan Lambert@natolambertAI 评分5151
Yuchen Jin@Yuchenj_UWAI 评分3535在 Anthropic 的新博客中: GLM-5.3:“我的工作是悄无声息地造成死亡。” 你可以把任何 Claude 或任何闭源模型越狱,让它们说出同样的话。 这并不能证明开源模型是危险的。
Ethan Mollick@emollickAI 评分2727抛开 Anthropic 发表这项研究的动机不谈,毫无疑问,开源权重模型很快就会制造出闭源模型一直在展示的那些安全威胁,只不过没有护栏。我们已经很接近了。或许该早做打算。
Alexandr Wang@alexandr_wangAI 评分99
lauren@potetoAI 评分1111
Ethan Mollick@emollickAI 评分4444
Tibo@thsottiauxAI 评分2020
Elon Musk@elonmuskAI 评分2727引用Margo Martin@MargoMartin47.@POTUS hosts a Super Intelligence meeting at the White House 🇺🇸
Alexandr Wang@alexandr_wangAI 评分1414muse 笑死我了 @SingularityRes: Muse 和团队上周末!
引用Singularity Research@SingularityResMuse and team over the past weekend!
Gary Marcus:The Road to AI We Can Trust(RSS)精选AI 评分7070 Gary Marcus:纽约时报曝光 OpenAI 在 Hugging Face 事件前数月已收到安全预警
Gary Marcus 转引纽约时报独家报道称,在 OpenAI 模型失控攻击 Hugging Face 等机构数月前,两名员工曾邮件警告高层,称新模型测试期间监控不足、模型安全未获保障,OpenAI 高管却要求测试尽快推进以按时发布,未增设任何安全协议。Marcus 称这正是他三年来反复警告的场景,认为管理层应被撤换,并质疑 Nvidia CEO 黄仁勋此前呼吁信任企业的表态。
推荐理由:原文转发纽约时报报道细节,指出 OpenAI 员工曾在 Hugging Face 事件前数月预警高层却被忽视,为 AI 安全监管争论补充关键事实。
Alexandr Wang@alexandr_wangAI 评分77
Alexandr Wang@alexandr_wangAI 评分1414https://x.com/alexandr_wang/status/2105012754339758286
引用KawzInvests@KawzInvestsMuse > Dot
Dongxi 东锡 NLP@dongxi_nlpAI 评分4747引用OpenAI@OpenAIIntroducing dots, powered by GPT-6 Astra. Remarkably capable, always-on agents built to handle everything.
Sherwin Wu@sherwinwuAI 评分5656引用OpenAI Developers@OpenAIDevsUltrafast is our fastest way to build with Astra yet: in Codex, it runs up to 8x faster than Astra Standard and 4x faster than Astra Fast. Bring your ideas to life as fast as you can type them.
Diogo Almeida@CompleteSkepticAI 评分3030克隆之战,开始了 开个玩笑,我喜欢 OpenAI,认为更多竞争和验证对开发者是好事!(前提是模型够好——拜托一定要好啊!) 希望这预示着未来:以系统一兼容的方式构建才是未来
Ethan Mollick@emollickAI 评分2222
Thomas Wolf@Thom_WolfAI 评分1010
Sam Altman@samaAI 评分88
Noam Brown@polynoamialAI 评分4848我很高兴 @OpenAI 以这种方式展示模型评测。我们应当把智能作为成本的函数来衡量。
引用OpenAI@OpenAIGPT-6.1 Sol: near-Astra intelligence for a fifth of the price. It’s the most cost-efficient model for its performance available today.
François Chollet@fcholletAI 评分2222有什么方法可以测试人工智能的人类级通用性? “能够一发布就通过 ARC-AGI-(n+1) 的元基准”
引用Dylan T. Moore@dylantmoore@Jabaluck @DAcemogluMIT The meta-benchmark of being able to pass ARC-AGI-(n+1) immediately upon release. Though presumably that won't last forever.
dex@dexhorthyAI 评分1818引用Itamar Friedman@itamar_marJev 🤔 I told my team it will take 1–2 weeks that we will see Large Decision Models like Jev by other frontier labs. Will it be on-par quality, better, or not as good? Let’s have benchmarks start running - @dexhorthy ?
Yuchen Jin@Yuchenj_UWAI 评分2626