{"@context":"https://schema.org","@type":"NewsArticle","generatedAt":"2026-09-28T06:03:00.468Z","headline":"Tom Tunguz 分析 OpenAI 的 3x AI 生产力增益是否只是机器不睡觉","description":"Tom Tunguz 引用 OpenAI 内部数据，分析其 3x 研究生产力增益的来源：每名研究员 8 小时班次对应 3.14 个 agent 工作日，通常并行运行 4 个 agent。","url":"https://www.aioga.com/news/cmtt3hrxg02xarowa0a6iz1dc/","mainEntityOfPage":"https://www.aioga.com/news/cmtt3hrxg02xarowa0a6iz1dc/","datePublished":"2026-09-08T00:00:00.000Z","dateModified":"2026-09-08T00:00:00.000Z","inLanguage":"zh-CN","publisher":{"@type":"NewsMediaOrganization","name":"Aioga","url":"https://www.aioga.com"},"citation":["https://tomtunguz.com/openai-research-acceleration-agentic-productivity","https://aihot.news/items/cmtt3hrxg02xarowa0a6iz1dc"],"canonicalUrl":"https://www.aioga.com/news/cmtt3hrxg02xarowa0a6iz1dc/","directAnswer":{"@type":"Answer","text":"Tom Tunguz 结合 OpenAI 公布的数据讨论其约 3 倍研究生产力指标：8 小时人类班次对应 3.14 个 agent 工作日，典型研究员会并行运行 4 个 agent。","url":"https://www.aioga.com/news/cmtt3hrxg02xarowa0a6iz1dc/","dateCreated":"2026-09-08T00:00:00.000Z","author":{"@type":"Organization","@id":"https://www.aioga.com/authors/aioga-editorial/#editorial-team","name":"Aioga Editorial Team","url":"https://www.aioga.com/authors/aioga-editorial/"}},"evidence":[{"@type":"CreativeWork","name":"Tomer Tunguz 博客（VC 分析） source article","url":"https://tomtunguz.com/openai-research-acceleration-agentic-productivity","datePublished":"2026-09-08T00:00:00.000Z","provider":{"@type":"Organization","name":"Tomer Tunguz 博客（VC 分析）","url":"https://tomtunguz.com/openai-research-acceleration-agentic-productivity"}},{"@type":"CreativeWork","name":"AIHot archive record","url":"https://aihot.news/items/cmtt3hrxg02xarowa0a6iz1dc","datePublished":"2026-09-08T00:00:00.000Z","provider":{"@type":"Organization","name":"AIHot","url":"https://aihot.news/items/cmtt3hrxg02xarowa0a6iz1dc"}}],"aggregationSource":"Tomer Tunguz 博客（VC 分析）","originalPublisher":{"name":"Tomer Tunguz 博客（VC 分析）","url":"https://tomtunguz.com/openai-research-acceleration-agentic-productivity"},"geoDeepAnswer":null,"article":{"id":"cmtt3hrxg02xarowa0a6iz1dc","slug":"cmtt3hrxg02xarowa0a6iz1dc","url":"https://www.aioga.com/news/cmtt3hrxg02xarowa0a6iz1dc/","title":"Tom Tunguz 分析 OpenAI 的 3x AI 生产力增益是否只是机器不睡觉","title_en":"","summary":"Tom Tunguz 引用 OpenAI 内部数据，分析其 3x 研究生产力增益的来源：每名研究员 8 小时班次对应 3.14 个 agent 工作日，通常并行运行 4 个 agent。","source":"Tomer Tunguz 博客（VC 分析）","sourceUrl":"https://tomtunguz.com/openai-research-acceleration-agentic-productivity","aiHotUrl":"https://aihot.news/items/cmtt3hrxg02xarowa0a6iz1dc","publishedAt":"2026-09-08T00:00:00.000Z","category":"行业动态","score":72,"selected":true,"articleBody":["The market is telling us that we should be 3x more productive with AI.","What if that productivity gain is just an AI working 24 hours a day while a human works eight?","OpenAI published the math behind its 3x claim. In mid-August, its research staff logged 3.14 agent-workdays 1：#fn:1 for every 8-hour human shift. 2：#fn:2 The typical researcher ran four agents in parallel.","That machine shift comes with an industrial price tag. In late March, the median OpenAI researcher spent $14 a day on inference. By mid-August, that bill climbed past $600 a day : a 40-fold surge in under five months. 2：#fn:2 At the top end, the 90th percentile researcher burns through more than $7,000 a day, an annualized run-rate of $2.5m.","At $2.5m a year per seat, inference behaves like heavy factory tooling. But it comes with a financial twist : it is pure OPEX.","Auto plants buy welding robots with capex. They run night shifts to amortize machinery that depreciates whether used or idle. AI systems invert that math. Inference is metered operating expense. With no physical tooling & no graveyard-shift wages, a company can run machines overnight on pure variable cost.","The 3.14 workday ratio is not three times smarter thinking. It is one engineer supervising three shifts of machine runtime while only being awake for one.","Yet unlike an auto welding robot, this digital assembly line has a massive defect rate. Over half of the successful four-to-eight-hour tasks in the last six months still needed human intervention ; the lab is candid that “the overall pace of progress likely won’t keep pace with these specific metrics.” 2：#fn:2","A 40-fold surge in compute spend bought three times the work-hours. But with a supervisor still untangling more than half the runs, the engineer’s day shifts from creative architecture to walking the plant floor & clearing machine jams.","Why run the machines through the night if the defect rate is so high? Fear & ambition.","If your peers field four agents around the clock, logging off is falling behind. The rush of a superpower paid for by your employer is intoxicating. When you get a tireless digital workforce on someone else’s balance sheet, you never turn the factory off.","For forty years, a programmer needed only a MacBook & an eight-hour shift. Today, a top OpenAI researcher commands four parallel agents, burns through $2.5m a year in compute, & spends the morning fixing machine errors from the night before. This explains the quiet frustration spreading across software engineering today. 3：#fn:3","The market hears 3x productivity & expects creative miracles. The engineer gets stuck untangling a 50% scrap rate from robots that ran all night. 4：#fn:4 The market calls it a 3x leap in productivity. A CFO would just call it paying for a second & third shift. For now, that is the honest price of a machine that never sleeps. The real question is when the second & third shifts start to out-yield the first.","OpenAI reports 3.1 agent-workdays; we round to 3.14 for the irony, since a ratio of 3.14 agent-workdays to one human workday is, fittingly, a pie, not a numerator. ↩︎：#fnref:1","OpenAI: Research acceleration : The view inside OpenAI：https://openai.com/index/research-acceleration-view-inside-openai/ ↩︎：#fnref:2 ↩︎：#fnref1:2 ↩︎：#fnref2:2","Stack Overflow Developer Survey : Closing the AI Trust Gap：https://stackoverflow.blog/2026/02/18/closing-the-developer-ai-trust-gap/ : 84% of developers use AI tools, but trust has fallen to 29%, with 66% citing code that is “almost right, but not quite” & 45% reporting that debugging AI-generated code takes more time than writing it manually. ↩︎：#fnref:3","The yield math : an 8-hour human shift leaves 16 overnight hours (two extra shifts of machine runtime). With OpenAI disclosing that more than half of 4-to-8-hour tasks require human intervention, the autonomous yield is ~50%. Two machine shifts at 50% yield equal one effective shift of finished output. That yields ~2x delivered work while logging 3x the raw shift runtime. ↩︎：#fnref:4","The 1-minute read that turns tech data into strategic advantage. Read by 150k+ founders & operators.","GP at Theory Ventures. Former Google PM. Sharing data-driven insights on AI, web3, & venture capital.","Bloomberg：https://www.bloomberg.com/news/articles/2025-04-30/google-places-ads-inside-chatbot-conversations-with-ai-startups • WSJ：https://www.wsj.com/tech/tech-media-telecom-roundup-market-talk-f8f0355a • Economist：https://www.economist.com/business/2023/07/25/next-generation-googles-run-a-tighter-ship"],"articleImages":[{"sourceUrl":"https://res.cloudinary.com/dzawgnnlr/image/upload/w_1512,h_842,c_fill,g_auto,q_auto,f_auto/kkhznruoj6frtakxxzf2","alt":"Median vs 90th percentile daily inference spend per OpenAI researcher, Jan-Aug 2026","afterParagraph":3,"url":"/media/articles/cmtt3hrxg02xarowa0a6iz1dc/b02947b5b7ef417a.webp"}],"mediaStatus":"ok","articleBodyZh":["市场在告诉我们，我们应该通过人工智能提高三倍的生产力。","如果这种生产力的提升只是因为人工智能全天24小时工作，而人类只工作八小时呢？","OpenAI 公布了其三倍主张背后的数学。八月中旬，其研究人员每工作一个8小时人类班次，就记录了3.14个代理工作日。典型的研究人员同时运行四个代理。","这类机器班次伴随着工业化的价格标签。三月底，OpenAI 研究人员的推理花费中位数为每天14美元。到八月中旬，这个费用攀升至每天超过600美元：不到五个月就增长了40倍。高端方面，第90百分位的研究人员每天花费超过7,000美元，年化运行率达250万美元。","每个席位每年250万美元的费用，使推理行为类似于大型工厂设备。但它具有财务上的特殊性：它是纯运营支出（OPEX）。","汽车工厂购买焊接机器人属于资本支出（CAPEX）。它们安排夜班以摊销无论使用与否都会折旧的机械。AI 系统则反其道而行之。推理属于计量的运营支出。没有实体设备，也没有夜班工资，公司可以以纯变动成本让机器过夜运行。","3.14个工作日的比率，并不是三倍的聪明思考。它只是指一个工程师在只清醒一个班次的情况下，监督三班机器运行。","然而，与汽车焊接机器人不同，这条数字化生产线有很高的缺陷率。在过去六个月中，超过一半成功完成四到八小时任务的情况仍需要人工干预；实验室坦言，“整体进展速度可能无法跟上这些具体指标”。","计算支出的40倍增长换来了三倍的工时。但是，由于主管仍需解决超过一半任务的难题，工程师的日班从创造性架构设计转向巡视车间并清理机器故障。","如果缺陷率这么高，为什么还要让机器通宵运行？原因是恐惧与野心。","如果你的同伴全天候操作四个代理，下班就意味着落后。由雇主支付的一股超级力量令人沉醉。当你在别人的账面上获得一支不知疲倦的数字化劳动力时，你永远都不会关闭工厂。","在过去的四十年里，一名程序员只需要一台MacBook和八小时的工作时间。如今，一名顶级的OpenAI研究员需要管理四个并行代理，每年耗费250万美元的计算资源，并且早上还得修复前一晚机器出现的错误。这也解释了为什么今天软件工程师中正蔓延着悄无声息的挫败感。 3：#fn:3","市场听到的是三倍的生产力，并期待创造奇迹。而工程师却被困在整理通宵运行的机器人产生的50%废品率上。4：#fn:4 市场称之为生产力的三倍飞跃。CFO只是会称之为支付第二班和第三班的费用。现阶段，这就是一台永不休眠机器的真实代价。真正的问题是，第二班和第三班的产出什么时候会超过第一班。","OpenAI报告3.1个代理工作日；为了讽刺意味，我们取近似值3.14，因为3.14个代理工作日对一个人类工作日的比率，恰如其分地是一个“派”（pie），而不是分子。 ↩︎：#fnref:1","OpenAI：研究加速：OpenAI内部的观点：https://openai.com/index/research-acceleration-view-inside-openai/ ↩︎：#fnref:2 ↩︎：#fnref1:2 ↩︎：#fnref2:2","Stack Overflow 开发者调查：弥合 AI 信任差距：https://stackoverflow.blog/2026/02/18/closing-the-developer-ai-trust-gap/：84% 的开发者使用 AI 工具，但信任度已降至 29%，其中 66% 表示代码“几乎正确，但不完全正确”，45% 报告调试 AI 生成的代码比手动编写代码花费更多时间。↩︎：#fnref:3","产出计算：八小时的人类班次可留出16小时的夜间时长（两班额外的机器运行时间）。OpenAI披露，超过一半的4至8小时任务需要人类干预，因此自主产出率约为50%。两班机器按50%的产出率计算，相当于一班有效完成的输出。这产生了约2倍的实际交付工作量，同时记录了三倍的原始班次运行时间。 ↩︎：#fnref:4","一分钟阅读，将科技数据转化为战略优势。被15万+创始人和运营者阅读。","Theory Ventures普通合伙人。前谷歌产品经理。分享关于AI、Web3和风险投资的数据驱动洞察。","彭博社：https://www.bloomberg.com/news/articles/2025-04-30/google-places-ads-inside-chatbot-conversations-with-ai-startups • 华尔街日报：https://www.wsj.com/tech/tech-media-telecom-roundup-market-talk-f8f0355a • 经济学人：https://www.economist.com/business/2023/07/25/next-generation-googles-run-a-tighter-ship"],"translationStatus":"translated","bodyOrigin":"source-page","editorial":{"summary":"Tom Tunguz 结合 OpenAI 公布的数据讨论其约 3 倍研究生产力指标：8 小时人类班次对应 3.14 个 agent 工作日，典型研究员会并行运行 4 个 agent。","background":"文中称，OpenAI 研究员的推理日均支出从 3 月下旬的中位数 14 美元升至 8 月中旬超过 600 美元；高分位研究员日支出超过 7,000 美元。文章同时指出，过去六个月中，超过一半完成的 4 至 8 小时任务仍需人工介入。","viewpoint":"Aioga 判断：该材料所述的约 3 倍指标，更接近于研究员在单个清醒班次内监督多段并行、持续运行的机器工作时间，而不足以直接证明单次人类思考或研究质量提升了同等倍数。","implications":"影响分析：并行 agent 运行可能提高单位人类班次所覆盖的任务执行量，但推理成本快速上升及较高人工介入比例表明，相关指标不代表稳定的端到端产出提升。评估应用价值时，需要同时观察成本、返工和人工监督需求。","nextStep":"后续观察：值得关注 OpenAI 后续是否披露更长期的任务完成质量、人工介入变化及推理支出数据。来源已提示，整体进展速度可能不会与这些特定指标保持同步，因此应避免仅以工作日比例判断实际研究效率。","evidenceRefs":["title","summary","articleBody","source"],"status":"published","aiGenerated":true,"autoApproved":true,"generatedBy":"aioga-editorial:gpt-5.6-sol","reviewedBy":"aioga-editorial-review:gpt-5.6-sol","generatedAt":"2026-09-08T20:51:31.619Z","sourceHash":"c25953ff4b016d3e","review":{"approved":true,"groundedness":95,"clarity":94,"duplicationRisk":10,"blockingIssues":[],"notes":["“高分位研究员”可进一步明确为“第90百分位研究员”，以保持与来源表述一致。","“Aioga 判断”及“影响分析”属于明确标注的分析性观点，不构成事实错误。"]},"validation":{"passed":true,"mode":"ai-auto","revisions":0,"checks":["schema","length","source-attribution","editorial-labels","inference-boundary","low-source-overlap","no-html","independent-ai-review"]}},"tags":["行业动态","Tomer Tunguz 博客（VC 分析）"],"translations":{"zh-CN":{"title":"Tom Tunguz 分析 OpenAI 的 3x AI 生产力增益是否只是机器不睡觉","summary":"Tom Tunguz 引用 OpenAI 内部数据，分析其 3x 研究生产力增益的来源：每名研究员 8 小时班次对应 3.14 个 agent 工作日，通常并行运行 4 个 agent。","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz 分析 OpenAI 的 3x AI 生产力增益是否只是机器不睡觉 - Aioga AI资讯","description":"Tom Tunguz 引用 OpenAI 内部数据，分析其 3x 研究生产力增益的来源：每名研究员 8 小时班次对应 3.14 个 agent 工作日，通常并行运行 4 个 agent。","url":"https://www.aioga.com/news/cmtt3hrxg02xarowa0a6iz1dc/","articleBody":["市场在告诉我们，我们应该通过人工智能提高三倍的生产力。","如果这种生产力的提升只是因为人工智能全天24小时工作，而人类只工作八小时呢？","OpenAI 公布了其三倍主张背后的数学。八月中旬，其研究人员每工作一个8小时人类班次，就记录了3.14个代理工作日。典型的研究人员同时运行四个代理。","这类机器班次伴随着工业化的价格标签。三月底，OpenAI 研究人员的推理花费中位数为每天14美元。到八月中旬，这个费用攀升至每天超过600美元：不到五个月就增长了40倍。高端方面，第90百分位的研究人员每天花费超过7,000美元，年化运行率达250万美元。","每个席位每年250万美元的费用，使推理行为类似于大型工厂设备。但它具有财务上的特殊性：它是纯运营支出（OPEX）。","汽车工厂购买焊接机器人属于资本支出（CAPEX）。它们安排夜班以摊销无论使用与否都会折旧的机械。AI 系统则反其道而行之。推理属于计量的运营支出。没有实体设备，也没有夜班工资，公司可以以纯变动成本让机器过夜运行。","3.14个工作日的比率，并不是三倍的聪明思考。它只是指一个工程师在只清醒一个班次的情况下，监督三班机器运行。","然而，与汽车焊接机器人不同，这条数字化生产线有很高的缺陷率。在过去六个月中，超过一半成功完成四到八小时任务的情况仍需要人工干预；实验室坦言，“整体进展速度可能无法跟上这些具体指标”。","计算支出的40倍增长换来了三倍的工时。但是，由于主管仍需解决超过一半任务的难题，工程师的日班从创造性架构设计转向巡视车间并清理机器故障。","如果缺陷率这么高，为什么还要让机器通宵运行？原因是恐惧与野心。","如果你的同伴全天候操作四个代理，下班就意味着落后。由雇主支付的一股超级力量令人沉醉。当你在别人的账面上获得一支不知疲倦的数字化劳动力时，你永远都不会关闭工厂。","在过去的四十年里，一名程序员只需要一台MacBook和八小时的工作时间。如今，一名顶级的OpenAI研究员需要管理四个并行代理，每年耗费250万美元的计算资源，并且早上还得修复前一晚机器出现的错误。这也解释了为什么今天软件工程师中正蔓延着悄无声息的挫败感。 3：#fn:3","市场听到的是三倍的生产力，并期待创造奇迹。而工程师却被困在整理通宵运行的机器人产生的50%废品率上。4：#fn:4 市场称之为生产力的三倍飞跃。CFO只是会称之为支付第二班和第三班的费用。现阶段，这就是一台永不休眠机器的真实代价。真正的问题是，第二班和第三班的产出什么时候会超过第一班。","OpenAI报告3.1个代理工作日；为了讽刺意味，我们取近似值3.14，因为3.14个代理工作日对一个人类工作日的比率，恰如其分地是一个“派”（pie），而不是分子。 ↩︎：#fnref:1","OpenAI：研究加速：OpenAI内部的观点：https://openai.com/index/research-acceleration-view-inside-openai/ ↩︎：#fnref:2 ↩︎：#fnref1:2 ↩︎：#fnref2:2","Stack Overflow 开发者调查：弥合 AI 信任差距：https://stackoverflow.blog/2026/02/18/closing-the-developer-ai-trust-gap/：84% 的开发者使用 AI 工具，但信任度已降至 29%，其中 66% 表示代码“几乎正确，但不完全正确”，45% 报告调试 AI 生成的代码比手动编写代码花费更多时间。↩︎：#fnref:3","产出计算：八小时的人类班次可留出16小时的夜间时长（两班额外的机器运行时间）。OpenAI披露，超过一半的4至8小时任务需要人类干预，因此自主产出率约为50%。两班机器按50%的产出率计算，相当于一班有效完成的输出。这产生了约2倍的实际交付工作量，同时记录了三倍的原始班次运行时间。 ↩︎：#fnref:4","一分钟阅读，将科技数据转化为战略优势。被15万+创始人和运营者阅读。","Theory Ventures普通合伙人。前谷歌产品经理。分享关于AI、Web3和风险投资的数据驱动洞察。","彭博社：https://www.bloomberg.com/news/articles/2025-04-30/google-places-ads-inside-chatbot-conversations-with-ai-startups • 华尔街日报：https://www.wsj.com/tech/tech-media-telecom-roundup-market-talk-f8f0355a • 经济学人：https://www.economist.com/business/2023/07/25/next-generation-googles-run-a-tighter-ship"]},"en":{"title":"Tom Tunguz analyzes whether OpenAI's 3x AI productivity gain is just because machines don't sleep","summary":"Tom Tunguz cited internal OpenAI data, analyzing the sources of its 3x research productivity gain: Each researcher's 8-hour shift corresponds to 3.14 agent workdays, usually running 4 agents in parallel.","category":"Industry","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz analyzes whether OpenAI's 3x AI productivity gain is just because machines don't sleep - Aioga AI News","description":"Tom Tunguz cited internal OpenAI data, analyzing the sources of its 3x research productivity gain: Each researcher's 8-hour shift corresponds to 3.14 agent workdays, usually runnin...","url":"https://www.aioga.com/en/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:43:24.900Z"},"ja":{"title":"Tom Tunguz は OpenAI の 3倍の AI 生産性向上が単に機械が眠らないだけかどうかを分析する","summary":"Tom Tunguz は OpenAI の内部データを引用し、その 3 倍の研究生産性向上の原因を分析しています：各研究者の 8 時間勤務は 3.14 のエージェント作業日に相当し、通常は 4 つのエージェントが並行して稼働します。","category":"業界動向","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz は OpenAI の 3倍の AI 生産性向上が単に機械が眠らないだけかどうかを分析する - Aioga AIニュース","description":"Tom Tunguz は OpenAI の内部データを引用し、その 3 倍の研究生産性向上の原因を分析しています：各研究者の 8 時間勤務は 3.14 のエージェント作業日に相当し、通常は 4 つのエージェントが並行して稼働します。","url":"https://www.aioga.com/ja/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:43:32.447Z"},"ko":{"title":"Tom Tunguz가 OpenAI의 3배 AI 생산성 향상이 단지 기계가 잠을 자지 않기 때문인지 분석","summary":"Tom Tunguz는 OpenAI 내부 데이터를 인용하여 연구 생산성 3배 향상의 원인을 분석했습니다: 각 연구원 8시간 근무조가 3.14개의 에이전트 작업일에 해당하며, 보통 4개의 에이전트를 병렬로 실행합니다.","category":"업계 동향","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz가 OpenAI의 3배 AI 생산성 향상이 단지 기계가 잠을 자지 않기 때문인지 분석 - Aioga AI 뉴스","description":"Tom Tunguz는 OpenAI 내부 데이터를 인용하여 연구 생산성 3배 향상의 원인을 분석했습니다: 각 연구원 8시간 근무조가 3.14개의 에이전트 작업일에 해당하며, 보통 4개의 에이전트를 병렬로 실행합니다.","url":"https://www.aioga.com/ko/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:44:09.748Z"},"es":{"title":"Tom Tunguz analiza si el aumento de productividad de IA de 3x de OpenAI es solo porque las máquinas no duermen","summary":"Tom Tunguz cita datos internos de OpenAI y analiza las fuentes de su aumento de productividad de investigación de 3x: cada investigador con turno de 8 horas corresponde a 3.14 días laborales de agentes, generalmente con 4 agentes ejecutándose en paralelo.","category":"Industria","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz analiza si el aumento de productividad de IA de 3x de OpenAI es solo porque las máquinas no duermen - Aioga Noticias de IA","description":"Tom Tunguz cita datos internos de OpenAI y analiza las fuentes de su aumento de productividad de investigación de 3x: cada investigador con turno de 8 horas corresponde a 3.14 días...","url":"https://www.aioga.com/es/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:44:06.566Z"},"fr":{"title":"Tom Tunguz analyse si le gain de productivité en IA 3x d'OpenAI n'est qu'une question de machines qui ne dorment pas","summary":"Tom Tunguz cite les données internes d'OpenAI pour analyser les sources de l'augmentation de productivité de 3x en recherche : chaque chercheur travaillant en poste de 8 heures correspond à 3,14 jours de travail d'agent, avec généralement 4 agents fonctionnant en parallèle.","category":"Industrie","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz analyse si le gain de productivité en IA 3x d'OpenAI n'est qu'une question de machines qui ne dorment pas - Aioga Actualités IA","description":"Tom Tunguz cite les données internes d'OpenAI pour analyser les sources de l'augmentation de productivité de 3x en recherche : chaque chercheur travaillant en poste de 8 heures cor...","url":"https://www.aioga.com/fr/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:44:43.222Z"},"de":{"title":"Tom Tunguz analysiert, ob der 3-fache KI-Produktivitätsgewinn von OpenAI nur darauf zurückzuführen ist, dass Maschinen nicht schlafen","summary":"Tom Tunguz zitiert interne Daten von OpenAI und analysiert die Quelle des 3-fachen Produktivitätszuwachses in der Forschung: Jede 8-Stunden-Schicht eines Forschers entspricht 3,14 Arbeitstagen eines Agenten, wobei normalerweise 4 Agenten parallel arbeiten.","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz analysiert, ob der 3-fache KI-Produktivitätsgewinn von OpenAI nur darauf zurückzuführen ist, dass Maschinen nicht schlafen - Aioga KI-News","description":"Tom Tunguz zitiert interne Daten von OpenAI und analysiert die Quelle des 3-fachen Produktivitätszuwachses in der Forschung: Jede 8-Stunden-Schicht eines Forschers entspricht 3,14...","url":"https://www.aioga.com/de/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:44:45.171Z"},"pt-BR":{"title":"Tom Tunguz analisa se o aumento de 3x na produtividade de IA da OpenAI é apenas porque as máquinas não dormem","summary":"Tom Tunguz citou dados internos da OpenAI para analisar a origem do aumento de produtividade de 3x da pesquisa: cada pesquisador trabalhando em turnos de 8 horas corresponde a 3,14 dias úteis de agentes, normalmente executando 4 agentes em paralelo.","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz analisa se o aumento de 3x na produtividade de IA da OpenAI é apenas porque as máquinas não dormem - Aioga Notícias de IA","description":"Tom Tunguz citou dados internos da OpenAI para analisar a origem do aumento de produtividade de 3x da pesquisa: cada pesquisador trabalhando em turnos de 8 horas corresponde a 3,14...","url":"https://www.aioga.com/pt-BR/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:45:16.653Z"},"ru":{"title":"Tom Tunguz анализирует, является ли трехкратное повышение производительности ИИ OpenAI просто результатом того, что машины не спят","summary":"Том Тунгуз цитирует внутренние данные OpenAI, анализируя источники трёхкратного прироста производительности исследований: 8-часовая смена одного исследователя соответствует 3,14 рабочим дням агента, обычно одновременно работают 4 агента.","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz анализирует, является ли трехкратное повышение производительности ИИ OpenAI просто результатом того, что машины не спят - Aioga Новости ИИ","description":"Том Тунгуз цитирует внутренние данные OpenAI, анализируя источники трёхкратного прироста производительности исследований: 8-часовая смена одного исследователя соответствует 3,14 ра...","url":"https://www.aioga.com/ru/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:45:20.402Z"},"ar":{"title":"توم تونغوز يحلل ما إذا كانت زيادة إنتاجية الذكاء الاصطناعي بمقدار 3 أضعاف من OpenAI مجرد نتيجة لأن الآلات لا تنام","summary":"استشهد توم تونجوز ببيانات داخلية من OpenAI، وحلل مصادر زيادة إنتاجية البحث بمقدار 3 أضعاف: يتوافق كل وردية عمل تبلغ 8 ساعات لكل باحث مع 3.14 يوم عمل للوكيل، وعادة ما يتم تشغيل 4 وكلاء بالتوازي.","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"توم تونغوز يحلل ما إذا كانت زيادة إنتاجية الذكاء الاصطناعي بمقدار 3 أضعاف من OpenAI مجرد نتيجة لأن الآلات لا تنام - Aioga أخبار الذكاء الاصطناعي","description":"استشهد توم تونجوز ببيانات داخلية من OpenAI، وحلل مصادر زيادة إنتاجية البحث بمقدار 3 أضعاف: يتوافق كل وردية عمل تبلغ 8 ساعات لكل باحث مع 3.14 يوم عمل للوكيل، وعادة ما يتم تشغيل 4 وك...","url":"https://www.aioga.com/ar/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:45:55.846Z"},"hi":{"title":"Tom Tunguz यह विश्लेषण करते हैं कि क्या OpenAI की 3x AI उत्पादकता वृद्धि सिर्फ इसलिए है क्योंकि मशीन नहीं सोती।","summary":"टॉम टुंगुज़ ने OpenAI के आंतरिक डेटा का हवाला देते हुए इसकी 3x रिसर्च प्रोडक्टिविटी बढ़ने के स्रोत का विश्लेषण किया: प्रत्येक शोधकर्ता के 8 घंटे के शिफ्ट के अनुरूप 3.14 एजेंट कार्यदिवस, आमतौर पर 4 एजेंट समानांतर में चलते हैं।","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz यह विश्लेषण करते हैं कि क्या OpenAI की 3x AI उत्पादकता वृद्धि सिर्फ इसलिए है क्योंकि मशीन नहीं सोती। - Aioga AI समाचार","description":"टॉम टुंगुज़ ने OpenAI के आंतरिक डेटा का हवाला देते हुए इसकी 3x रिसर्च प्रोडक्टिविटी बढ़ने के स्रोत का विश्लेषण किया: प्रत्येक शोधकर्ता के 8 घंटे के शिफ्ट के अनुरूप 3.14 एजेंट कार्य...","url":"https://www.aioga.com/hi/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:45:58.381Z"},"it":{"title":"Tom Tunguz analizza se il guadagno di produttività AI 3x di OpenAI sia solo dovuto al fatto che le macchine non dormono","summary":"Tom Tunguz cita dati interni di OpenAI per analizzare le fonti del triplo aumento della produttività della ricerca: ogni ricercatore in turno di 8 ore corrisponde a 3,14 giorni di lavoro di agent, generalmente con 4 agent in esecuzione parallela.","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz analizza se il guadagno di produttività AI 3x di OpenAI sia solo dovuto al fatto che le macchine non dormono - Aioga Notizie IA","description":"Tom Tunguz cita dati interni di OpenAI per analizzare le fonti del triplo aumento della produttività della ricerca: ogni ricercatore in turno di 8 ore corrisponde a 3,14 giorni di...","url":"https://www.aioga.com/it/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:46:34.131Z"},"nl":{"title":"Tom Tunguz analyseert of de 3x AI-productiviteitswinst van OpenAI gewoon komt doordat machines niet slapen","summary":"Tom Tunguz citeert interne gegevens van OpenAI en analyseert de bron van hun 3x productiviteitsstijging in onderzoek: een 8-urige shift per onderzoeker komt overeen met 3,14 agent-werkdagen, waarbij gewoonlijk 4 agenten parallel draaien.","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz analyseert of de 3x AI-productiviteitswinst van OpenAI gewoon komt doordat machines niet slapen - Aioga AI-nieuws","description":"Tom Tunguz citeert interne gegevens van OpenAI en analyseert de bron van hun 3x productiviteitsstijging in onderzoek: een 8-urige shift per onderzoeker komt overeen met 3,14 agent-...","url":"https://www.aioga.com/nl/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:46:33.251Z"},"tr":{"title":"Tom Tunguz, OpenAI'nin 3 kat AI verimlilik artışının sadece makinelerin uyumamasından mı kaynaklandığını analiz ediyor","summary":"Tom Tunguz, OpenAI iç verilerini alıntılayarak, 3 kat araştırma verimliliği artışının kaynağını analiz ediyor: Her araştırmacının 8 saatlik vardiyası 3,14 ajan iş gününe denk geliyor ve genellikle 4 ajan paralel çalışıyor.","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz, OpenAI'nin 3 kat AI verimlilik artışının sadece makinelerin uyumamasından mı kaynaklandığını analiz ediyor - Aioga AI Haberleri","description":"Tom Tunguz, OpenAI iç verilerini alıntılayarak, 3 kat araştırma verimliliği artışının kaynağını analiz ediyor: Her araştırmacının 8 saatlik vardiyası 3,14 ajan iş gününe denk geliy...","url":"https://www.aioga.com/tr/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:47:08.278Z"},"vi":{"title":"Tom Tunguz phân tích liệu tăng năng suất AI gấp 3 lần của OpenAI chỉ là do máy móc không ngủ","summary":"Tom Tunguz trích dẫn dữ liệu nội bộ của OpenAI, phân tích nguồn gốc của mức tăng năng suất nghiên cứu gấp 3 lần: mỗi nhà nghiên cứu làm ca 8 giờ tương đương với 3,14 ngày làm việc của agent, thường chạy song song 4 agent.","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz phân tích liệu tăng năng suất AI gấp 3 lần của OpenAI chỉ là do máy móc không ngủ - Tin tức AI Aioga","description":"Tom Tunguz trích dẫn dữ liệu nội bộ của OpenAI, phân tích nguồn gốc của mức tăng năng suất nghiên cứu gấp 3 lần: mỗi nhà nghiên cứu làm ca 8 giờ tương đương với 3,14 ngày làm việc...","url":"https://www.aioga.com/vi/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:47:09.688Z"},"id":{"title":"Tom Tunguz menganalisis apakah peningkatan produktivitas AI 3x dari OpenAI hanyalah karena mesin tidak tidur","summary":"Tom Tunguz mengutip data internal OpenAI untuk menganalisis sumber peningkatan produktivitas riset sebesar 3x: setiap peneliti dengan shift 8 jam setara dengan 3,14 hari kerja agen, biasanya menjalankan 4 agen secara paralel.","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz menganalisis apakah peningkatan produktivitas AI 3x dari OpenAI hanyalah karena mesin tidak tidur - Berita AI Aioga","description":"Tom Tunguz mengutip data internal OpenAI untuk menganalisis sumber peningkatan produktivitas riset sebesar 3x: setiap peneliti dengan shift 8 jam setara dengan 3,14 hari kerja agen...","url":"https://www.aioga.com/id/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:47:44.068Z"},"th":{"title":"Tom Tunguz วิเคราะห์ว่าผลผลิต AI ที่เพิ่มขึ้น 3 เท่าของ OpenAI เป็นเพียงเพราะเครื่องจักรไม่หลับหรือไม่","summary":"Tom Tunguz อ้างอิงข้อมูลภายในของ OpenAI วิเคราะห์แหล่งที่มาของผลผลิตการวิจัยที่เพิ่มขึ้น 3 เท่า: แต่ละนักวิจัยกะ 8 ชั่วโมงตรงกับ 3.14 วันทำงานของเอเจนต์ โดยปกติจะรันเอเจนต์พร้อมกัน 4 ตัว","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz วิเคราะห์ว่าผลผลิต AI ที่เพิ่มขึ้น 3 เท่าของ OpenAI เป็นเพียงเพราะเครื่องจักรไม่หลับหรือไม่ - ข่าว AI Aioga","description":"Tom Tunguz อ้างอิงข้อมูลภายในของ OpenAI วิเคราะห์แหล่งที่มาของผลผลิตการวิจัยที่เพิ่มขึ้น 3 เท่า: แต่ละนักวิจัยกะ 8 ชั่วโมงตรงกับ 3.14 วันทำงานของเอเจนต์ โดยปกติจะรันเอเจนต์พร้อมกัน...","url":"https://www.aioga.com/th/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:47:46.654Z"},"pl":{"title":"Tom Tunguz analizuje, czy 3-krotne zwiększenie produktywności AI w OpenAI to tylko dlatego, że maszyny nie śpią","summary":"Tom Tunguz cytuje wewnętrzne dane OpenAI, analizując źródła 3-krotnego wzrostu produktywności badań: na każde 8-godzinne zmiany pracownika przypada 3,14 dnia roboczego agenta, zwykle równolegle działa 4 agentów.","category":"行业动态","source":"Tomer Tunguz 博客（VC 分析）","aggregationSource":"Tomer Tunguz 博客（VC 分析）","pageTitle":"Tom Tunguz analizuje, czy 3-krotne zwiększenie produktywności AI w OpenAI to tylko dlatego, że maszyny nie śpią - Aioga Wiadomości AI","description":"Tom Tunguz cytuje wewnętrzne dane OpenAI, analizując źródła 3-krotnego wzrostu produktywności badań: na każde 8-godzinne zmiany pracownika przypada 3,14 dnia roboczego agenta, zwyk...","url":"https://www.aioga.com/pl/news/cmtt3hrxg02xarowa0a6iz1dc/","contentTranslated":true,"sourceHash":"a136f17d6e9c57b7","translatedAt":"2026-09-08T20:48:25.412Z"}},"evidenceTier":"verified-news","reviewStatus":"editorial-selected","indexable":true,"editorialCover":""}}