全部/科技/AI as Normal Technology · 最新文章技术博客与 newsletter feed · AI as Normal Technology · 最新文章59分钟前更新实时榜 20 条今日全部 20 条统计中今日全部:今天 00:00 起上过榜的每条热搜,含最高排名与在榜时长近 30 天历史展开HISTORY近 30 天历史柱高表示当天去重热搜数量← 更早 30 天更近 30 天 →实时榜收起2023202420252026202720282029203020312001200220032004200520062007200820092010211120122013201420152016201720182019202020今天10/23—11/21 有历史数据01The missing quadrant in responses to AIAnnouncing STABL: Societal Transformation and AI Bottlenecks LabSayash Kapoor02New paper: Emerging trends in AI agent evaluationUnderstanding the methods, questions, and ecosystem for evaluationSayash Kapoor03Steering AI Beyond AutomationA Research Agenda and Plan for ActionArvind Narayanan04A big-tent or small-tent AI safety movement?The unstated disagreement that underpins safety debatesArvind Narayanan05AI existential risk probabilities are (still) too unreliable to inform policyHow speculation gets laundered through pseudo-quantificationArvind Narayanan06The AI-as-Normal-Technology view of loss-of-control incidentsA middle ground between the cybersecurity and AI safety communitiesSayash Kapoor07AI agents can't yet do open-ended AI researchEarly evidence from two case studiesSayash Kapoor08What will be left for us to work on?My keynote at ICML 2026Arvind Narayanan09Up the Stack: How AI’s Escape From the Commodity Trap Risks Enterprise Lock-inCritics and boosters are both looking in the wrong placeArvind Narayanan10Why AI hasn’t replaced software engineers, and won’tCoding agents as normal technologyArvind Narayanan11Did Google’s AI agents really build an operating system for $916?The importance of independent evaluationSayash Kapoor12Do AI Risks Require Extraordinary Government Intervention?Let’s not skip the hard work of AI governanceSayash Kapoor13Open-world evaluations for measuring frontier AI capabilitiesIntroducing CRUX, a new project for evaluating AI on long, messy tasksSayash Kapoor14New Paper: Towards a science of AI agent reliabilityQuantifying the capability-reliability gapSayash Kapoor15AI Won’t Automatically Make Legal Services CheaperApplying the AI as Normal Technology framework to legal servicesJustin Curl16Fact checking Moravec's paradoxThis famous aphorism is neither true nor usefulArvind Narayanan17A guide to understanding AI as normal technologyAnd a big change for this newsletterArvind Narayanan18Could AI slow science?Confronting the production-progress paradoxSayash Kapoor19AGI is not a milestoneThere is no capability threshold that will lead to sudden impactsSayash Kapoor20AI as Normal TechnologyA new paper that we will expand into our next bookArvind Narayanan