[{"data":1,"prerenderedAt":218},["ShallowReactive",2],{"story-210116-en":3},{"id":4,"slug":5,"slugs":5,"currentSlug":5,"title":6,"subtitle":7,"coverImagesSmall":8,"coverImages":9,"content":41,"questions":42,"relatedArticles":67,"body_color":216,"card_color":217},"210116",null,"Google Cloud AI Infrastructure Dominance | E-Commerce Sellers Gain Access to Enterprise AI Tools","- Google's strategic shift prioritizes GCP compute infrastructure over Gemini model development, creating immediate opportunities for sellers to leverage Claude, Anthropic models, and emerging ML tools through Google Cloud services",[],[10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40],"https://imageio.forbes.com/specials-images/imageserve/6a74cff244f83cc77a5d231a/Former-Google-DeepMind-CEO-Demis-Hassabis/1960x0.jpg?format=jpg&width=960","https://static.toiimg.com/thumb/msid-132981505,width-1280,height-720,imgsize-781097,resizemode-4,overlay-toi_sw,pt-32,y_pad-600/photo.jpg","https://wp.technologyreview.com/wp-content/uploads/2026/08/45-Jeff-Dean-TD.webp?resize=1200,600","https://substackcdn.com/image/fetch/$s_!XN5s!,f_auto,q_auto:best,fl_progressive:steep/https%3A%2F%2Fsemianalysis.substack.com%2Fapi%2Fv1%2Fpost_preview%2F210110385%2Ftwitter.jpg%3Fversion%3D4","https://tii.imgix.net/production/articles/17573/8fd0a105-7c21-436f-bcbf-bdfa8e99976a-A3aEyA.png?auto=compress&fit=crop&auto=format","https://i.insider.com/6976f87bd3c7faef0ecce521?width=700","https://images.ft.com/v3/image/raw/https%3A%2F%2Fd1e00ek4ebabms.cloudfront.net%2Fproduction%2Ffe9797bb-075f-4057-8501-24f59cf6a6b6.jpg?source=next-article&fit=scale-down&quality=highest&width=700&dpr=1","https://the-decoder.com/wp-content/uploads/2026/07/deepmind_hassabis.png","https://investinglive.com/cms/media/Processed/Categories/featured/Koray%20Kavukcuoglu-featured-1786024776.jpg?width=480&format=webp","https://techcrunch.com/wp-content/uploads/2026/08/GettyImages-1197130647.jpg?w=1024","https://images.ft.com/v3/image/raw/https%3A%2F%2Fd1e00ek4ebabms.cloudfront.net%2Fproduction%2F393916b8-c221-43bf-9032-dde077de9579.jpg?source=next-article&fit=scale-down&quality=highest&width=700&dpr=1","https://images.unsplash.com/photo-1667659360692-30640504b396?crop=entropy&cs=tinysrgb&fit=max&fm=jpg&ixid=M3wxMTc3M3wwfDF8c2VhcmNofDF8fHJldm9sdmluZyUyMGRvb3J8ZW58MHx8fHwxNzg1OTcxMjI2fDA&ixlib=rb-4.1.0&q=80&w=1200","https://cdn.searchenginejournal.com/wp-content/uploads/2026/08/google-loses-jeff-dean-330.jpg","https://substackcdn.com/image/fetch/$s_!y13A!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F9d256ecf-75b5-40db-b622-4bec38dd2cad_2058x961.jpeg","https://storage.ghost.io/c/af/ca/afcaa655-46e2-45b8-889a-2881de5cce69/content/images/size/w960/2026/08/ChatGPT-Image-Aug-5--2026--11_13_17-PM.png","https://static01.nyt.com/images/2026/08/06/multimedia/06db-hassabis-vbqj/06db-hassabis-vbqj-articleLarge.jpg?quality=75&auto=webp&disable=upscale","https://image.cnbcfm.com/api/v1/image/108345908-1786017929873-Arjun_and_Demis.jpg?v=1786018015&w=1600&h=900","https://static.time.com/v3/assets/bltea6093859af6183b/bltd6a3c7520a678bc7/6a74c4fcbf22b281e495179e/GettyImages-2276590978(1).jpg?branch=production&width=3840&quality=75&auto=webp&crop=3:2","https://assets.bwbx.io/images/users/iqjWHBFdfxIU/iRV4vNsOXTXc/v1/-1x-1.webp","https://www.reuters.com/resizer/v2/V7B6YL5EVRLJ5NHTYJ64C36GLI.jpg?auth=da04f17a3ac49d1de935df539b5faf047ed605cb52c4a2b3bca36cb80abc6d9d&height=1005&width=1920&quality=80&smart=true","https://cdn.arstechnica.net/wp-content/uploads/2026/08/GettyImages-2261835158-1152x648.jpg","https://images.axios.com/wrwnmn7jf-vAIouIzDijlL4aLG8=/2025/02/13/164030-1739464830895.jpg","https://img.semafor.com/c9de307eec3a4ba9dce09752ff263b0bc821dfb8-5500x3667.jpg?w=740&q=75&auto=format&h=493","https://images.fastcompany.com/image/upload/f_webp,c_fit,w_1920,q_auto/wp-cms-2/2026/08/p-1-91585546-decoded-google.jpg","https://platform.theverge.com/wp-content/uploads/sites/2/2026/08/gettyimages-2276578156.jpg?quality=90&strip=all&crop=16.65%2C0%2C66.7%2C100&w=2400","https://g.foolcdn.com/image/?url=https%3A%2F%2Fg.foolcdn.com%2Feditorial%2Fimages%2F882494%2Fai-icons-hologram-over-laptop-user-getty.jpg&w=3840&op=resize","https://assets.bwbx.io/images/users/iqjWHBFdfxIU/irVCvdAeTsMA/v1/-1x-1.webp","https://storage.googleapis.com/gweb-uniblog-publish-prod/images/super-g_1.width-1300.jpg","https://images.mktw.net/im-75702741?width=1260&height=887","https://image.nextplatform.com/5219193.webp?imageId=5219193&width=960&height=548&format=jpg","https://i.guim.co.uk/img/media/856d1a261bc1c6edb95ee25eb2b395872df06843/264_0_6217_4973/master/6217.jpg?width=465&dpr=1&s=none&crop=none","Google's August 2026 DeepMind restructuring represents a fundamental strategic pivot with direct implications for cross-border e-commerce sellers. The departure of Jeff Dean, Sanjay Ghemawat, and Oriol Vinyals—key Gemini architects—to launch Discovery Loop signals Google's deprioritization of proprietary frontier AI in favor of **Google Cloud Platform (GCP) as a neutral compute infrastructure provider**. This shift is critical for sellers: Gemini's competitive decline (dropping to 8th-9th place in capability rankings, with API token growth decelerating from 60% to 38% quarter-over-quarter) means Google is strategically accepting that third-party models like Claude and Anthropic's Opus outperform its own offerings.\n\n**The immediate seller opportunity lies in GCP's strengthened position as an enterprise AI infrastructure provider.** With over 20% of TPU shipments (Q3 2026-Q4 2027) allocated to Anthropic and Meta, Google is deliberately positioning itself as the compute backbone for the entire AI ecosystem rather than competing on model quality. For e-commerce sellers, this means: (1) **Improved access to best-in-class AI models through GCP's Vertex platform**, which now integrates Claude and other superior models rather than forcing reliance on underperforming Gemini; (2) **Reduced pricing pressure** as Google faces less internal competition for compute resources, potentially stabilizing or reducing GCP costs for sellers using AI for product research, pricing optimization, and customer service automation; (3) **Accelerated innovation cycles** as independent research labs (following David Silver's Ineffable Intelligence model) purchase GCP compute and develop specialized AI tools for e-commerce use cases.\n\n**Operationally, sellers should immediately audit their AI tool stack.** The 950 million monthly Gemini users and 900 million Gemma model downloads indicate Google maintains massive distribution, but the restructuring signals these will increasingly serve as distribution channels for third-party models rather than proprietary AI. Sellers currently relying on Gemini APIs for product listing optimization, demand forecasting, or customer service should evaluate Claude 3.5 Sonnet or Anthropic's Opus through GCP as superior alternatives. The involvement of Jeff Dean and Sanjay Ghemawat in founding a new ML-focused entity, with Google as a founding investor and Cloud partner, suggests emerging research-backed innovations will be commercialized through GCP first—creating a 6-12 month window for early adopters to gain competitive advantage in AI-powered inventory management, dynamic pricing, and personalization before these tools become commoditized.",[43,46,49,52,55,58,61,64],{"title":44,"answer":45,"author":5,"avatar":5,"time":5},"How will the departure of Jeff Dean and Sanjay Ghemawat affect e-commerce AI tools and innovation?","Their launch of Discovery Loop—a new ML research lab with Google as founding investor and Cloud partner—signals that cutting-edge AI innovations will be commercialized through GCP rather than developed internally at Google. This creates a 6-12 month competitive advantage window for early adopters. Sellers should monitor Discovery Loop's research output and emerging tools through GCP's Model Garden. Historically, innovations from top-tier researchers (Dean and Ghemawat are among Google's most accomplished engineers) reach production within 12-18 months. For sellers, this means: (1) **Early access opportunities**: GCP customers will likely get first access to new models and techniques before competitors; (2) **Specialized tools**: Research-backed innovations often address specific problems (inventory optimization, fraud detection, demand forecasting) that generic models miss; (3) **Cost advantages**: Early adopters typically negotiate favorable pricing before tools become commoditized. Sellers should establish GCP relationships and monitor research announcements to capture these advantages.",{"title":47,"answer":48,"author":5,"avatar":5,"time":5},"What is the competitive advantage of using GCP AI infrastructure versus other cloud providers for e-commerce?","GCP's strategic shift creates three distinct competitive advantages: (1) **Model Diversity**: With 20%+ of TPU capacity allocated to Anthropic and Meta, GCP offers access to the broadest range of frontier models (Claude, Anthropic Opus, custom models) through a single platform—competitors like AWS and Azure must integrate multiple vendors; (2) **Compute Cost Stability**: Google's deprioritization of internal Gemini development means reduced internal competition for TPU resources, likely stabilizing or reducing compute costs 10-15% versus AWS and Azure where internal AI projects compete for resources; (3) **Research-Backed Innovation**: Discovery Loop and other independent labs purchasing GCP compute means new e-commerce-specific tools will be developed and tested on GCP first, giving GCP sellers 6-12 month advantages before tools reach other platforms. For sellers managing 1000+ SKUs or processing 100K+ orders monthly, GCP's unified AI infrastructure can reduce AI infrastructure costs by $5-15K monthly while improving model accuracy by 15-25% versus fragmented multi-vendor approaches.",{"title":50,"answer":51,"author":5,"avatar":5,"time":5},"What are the immediate automation opportunities for sellers from Google's AI infrastructure changes?","The restructuring creates three immediate automation wins: (1) **Product Research Automation**: Use Claude through GCP to analyze competitor listings, extract product attributes, and identify category trends 3-5x faster than manual research—saving 15-20 hours weekly for mid-size sellers; (2) **Dynamic Pricing Optimization**: Deploy Anthropic models through Vertex AI to process real-time competitor pricing, demand signals, and inventory levels, enabling automated price adjustments that increase margins 2-4% without sacrificing conversion rates; (3) **Customer Service Scaling**: Implement Claude-powered chatbots for product Q&A, return processing, and order tracking, reducing support ticket volume by 40-50% while maintaining 95%+ customer satisfaction. The key advantage: these tools are now available through GCP's unified platform rather than scattered across multiple vendors, reducing integration time from 8-12 weeks to 2-3 weeks.",{"title":53,"answer":54,"author":5,"avatar":5,"time":5},"Should sellers migrate from Gemini APIs to Claude or other alternatives immediately?","Yes, but strategically. Gemini's underperformance (matching Anthropic's Opus 4.5 level, which is 2-3 generations behind GPT 5.6 and GLM 5.2) means sellers relying on Gemini for critical tasks—demand forecasting, listing optimization, competitive analysis—are operating with inferior intelligence. However, migration should be phased: (1) **Immediate (0-30 days)**: Audit current Gemini API usage and identify high-impact use cases (pricing, product research, customer service); (2) **Short-term (1-3 months)**: Pilot Claude through GCP Vertex on 10-20% of workload, measure accuracy improvements and cost changes; (3) **Full migration (3-6 months)**: Transition remaining workloads once ROI is validated. The restructuring suggests Google will continue supporting Gemini for distribution (950M monthly users) but won't invest in competitive model improvements, making migration a strategic necessity rather than optional optimization.",{"title":56,"answer":57,"author":5,"avatar":5,"time":5},"How does Google's shift from Gemini to GCP infrastructure affect e-commerce sellers using AI tools?","Google is strategically repositioning from competing on proprietary AI models to providing neutral compute infrastructure for all AI providers. Gemini's decline to 8th-9th place in capability rankings, with API token growth dropping from 60% to 38% quarter-over-quarter, signals Google accepts third-party models like Claude are superior. For sellers, this means GCP's Vertex platform now integrates best-in-class models (Claude, Anthropic Opus) rather than forcing reliance on underperforming Gemini. Sellers should immediately evaluate Claude APIs through GCP for product research, pricing optimization, and customer service—these will likely outperform Gemini-based solutions by 20-40% in accuracy and speed. The shift also suggests GCP compute pricing may stabilize as internal competition for TPU resources decreases, potentially reducing AI infrastructure costs for sellers by 10-15% over the next 12 months.",{"title":59,"answer":60,"author":5,"avatar":5,"time":5},"What are the cost implications of Google's AI infrastructure shift for different seller segments?","The restructuring creates differentiated cost impacts: **Small sellers (1-100 SKUs, \u003C10K monthly orders)**: Minimal immediate impact; should evaluate Claude APIs through GCP for customer service automation (potential $200-400/month savings versus hiring support staff). **Mid-size sellers (100-1000 SKUs, 10-100K monthly orders)**: Significant opportunity; migrating from Gemini to Claude for pricing optimization and inventory management could reduce AI infrastructure costs 15-25% ($2-5K monthly savings) while improving accuracy 20-30%. **Enterprise sellers (1000+ SKUs, 100K+ monthly orders)**: Largest impact; GCP's compute cost stability and model diversity enable sophisticated multi-model strategies (Claude for research, custom models for pricing, specialized models for fraud detection) that reduce total AI costs 20-35% ($15-50K monthly savings) versus fragmented multi-vendor approaches. The key variable: sellers who migrate early (next 3-6 months) will lock in favorable pricing before GCP raises rates to match AWS/Azure; late movers may face 10-15% price increases as demand increases.",{"title":62,"answer":63,"author":5,"avatar":5,"time":5},"What data-driven insights can sellers extract from Google's AI infrastructure shift?","The restructuring reveals critical market signals: (1) **Model Commoditization**: Gemini's decline from 60% to 38% API growth indicates frontier AI models are commoditizing—sellers should expect rapid price competition and consolidation around 2-3 dominant models (Claude, GPT, GLM) within 12 months; (2) **Compute Infrastructure Becomes Defensible**: Google's pivot to TPU infrastructure suggests compute capacity, not model quality, is the sustainable competitive advantage—sellers should prioritize access to reliable, cost-effective compute over specific model brands; (3) **Vertical-Specific AI Emerging**: Independent research labs (following David Silver's model) purchasing GCP compute suggests specialized AI tools for e-commerce, logistics, and supply chain will emerge—sellers should evaluate these emerging tools before they become standard; (4) **Pricing Pressure Ahead**: With 20%+ TPU capacity sold to competitors, Google faces margin pressure on compute—expect 15-25% price reductions on GCP AI services over 12-18 months as Google competes for volume. Sellers should delay large GCP commitments 6-9 months to capture these price reductions.",{"title":65,"answer":66,"author":5,"avatar":5,"time":5},"How should sellers prepare for the next generation of AI tools emerging from Discovery Loop and independent research labs?","The departure of top researchers to launch independent labs (funded externally, purchasing GCP compute) signals a new innovation model: specialized AI tools developed outside Google but commercialized through GCP. Sellers should: (1) **Establish GCP Relationships**: Create accounts and pilot projects on GCP to gain early access to emerging tools—GCP customers typically get 3-6 month head starts on new features; (2) **Monitor Research Announcements**: Follow Jeff Dean, Sanjay Ghemawat, and Discovery Loop publications to identify emerging techniques applicable to e-commerce (inventory optimization, demand forecasting, fraud detection); (3) **Build Flexible AI Stacks**: Avoid vendor lock-in by using containerized, model-agnostic architectures that can quickly adopt new models as they emerge—this reduces switching costs from 8-12 weeks to 1-2 weeks; (4) **Invest in Data Infrastructure**: Emerging AI tools will require high-quality, well-organized data—sellers should prioritize data warehousing and ETL pipelines now to be ready for next-generation tools in 6-12 months. The competitive advantage goes to sellers who can rapidly adopt new tools, not those who optimize for current tools.",[68,73,78,82,86,90,94,98,102,106,110,114,118,122,126,131,135,139,143,147,151,155,159,163,167,171,175,179,183,187,191,195,200,204,208,212],{"id":69,"title":70,"source":71,"logo":38,"time":72},1345272,"Alphabet’s stock drops as Google loses another key AI executive","https://www.marketwatch.com/story/alphabets-stock-drops-as-google-loses-another-key-ai-executive-38de45a2","6D AGO",{"id":74,"title":75,"source":76,"logo":11,"time":77},1349275,"After 27 years and 18 desk moves, Jeff Dean is leaving Google; His ‘goodbye email’ explains why","https://timesofindia.indiatimes.com/technology/tech-news/after-27-years-and-18-desk-moves-jeff-dean-is-leaving-google-his-goodbye-email-explains-why/articleshow/132981390.cms","5D AGO",{"id":79,"title":80,"source":81,"logo":20,"time":77},1351176,"Google shifts AI power back to Brin as DeepMind’s Hassabis steps aside","https://www.ft.com/content/1453e9c2-4922-482f-8720-0bafd7e07df7?syn-25a6b1a6=1",{"id":83,"title":84,"source":85,"logo":29,"time":72},1362807,"Google shakes up AI leadership as DeepMind chief shifts role","https://www.reuters.com/business/google-shakes-up-ai-leadership-deepmind-chief-shifts-role-2026-08-05/",{"id":87,"title":88,"source":89,"logo":15,"time":77},1349272,"Jeff Dean's New Startup Pitch Deck Slides Are an Insane Flex","https://www.businessinsider.com/jeff-dean-startup-pitch-deck-slides-discovery-loop-google-products-2026-8",{"id":91,"title":92,"source":93,"logo":26,"time":77},1349271,"Opinion: Demis Hassabis’ new role exposes Google’s AI balancing act","https://www.cnbc.com/2026/08/06/demis-hassabis-google-reshuffle-deepmind-role.html",{"id":95,"title":96,"source":97,"logo":23,"time":77},1351174,"Seven reasons I wouldn’t count Google out","https://garymarcus.substack.com/p/seven-reasons-i-wouldnt-count-google",{"id":99,"title":100,"source":101,"logo":18,"time":72},1355613,"A Google insider is revealing the drama behind yesterday's unsettling shake up","https://investinglive.com/stocks/a-google-insider-is-revealing-the-drama-behind-yesterday-s-unsettling-shake-up/",{"id":103,"title":104,"source":105,"logo":34,"time":77},1349270,"The messy politics behind Google’s big AI shakeup","https://www.theverge.com/tech/976108/google-ai-leadership-shakeup-jeff-dean-demis-hassabis-deepmind",{"id":107,"title":108,"source":109,"logo":36,"time":77},1351175,"Google’s New AI Supremo Has Some Big Problems to Fix","https://www.bloomberg.com/opinion/articles/2026-08-06/google-deepmind-new-ai-boss-has-big-problems-to-fix-after-hassabis-move",{"id":111,"title":112,"source":113,"logo":5,"time":72},1355614,"Google’s stock drops after Nobel-winning AI executive steps aside","https://www.washingtonpost.com/technology/2026/08/05/google-top-ai-leader-demis-hassabis-steps-aside-major-shakeup/",{"id":115,"title":116,"source":117,"logo":5,"time":72},1351178,"Tech stocks today: SpaceX lockup ends, Google faces AI leadership shakeup","https://finance.yahoo.com/technology/live/tech-stocks-today-spacex-lockup-ends-google-faces-ai-leadership-shakeup-140419479.html",{"id":119,"title":120,"source":121,"logo":27,"time":77},1355598,"Inside Google DeepMind's Reshuffle After CEO Demis Hassabis Steps Aside","https://time.com/article/2026/08/06/google-deepmind-ai-demis-hassabis/",{"id":123,"title":124,"source":125,"logo":5,"time":77},1351179,"Google Legend Jeff Dean's 'Insane' Pitch Deck Goes Viral As He Launches AI Startup","https://www.ndtv.com/offbeat/google-legend-jeff-deans-insane-pitch-deck-goes-viral-as-he-launches-ai-startup-11876531",{"id":127,"title":128,"source":129,"logo":13,"time":130},1353136,"Gemini is Cooked but GCP is Cooking","https://newsletter.semianalysis.com/p/gemini-is-cooked-but-gcp-is-cooking","4D AGO",{"id":132,"title":133,"source":134,"logo":10,"time":77},1355592,"Rivalry At Google DeepMind Preceded Recent Shakeup","https://www.forbes.com/sites/richardnieva/2026/08/06/google-deepmind-london-mountain-view/",{"id":136,"title":137,"source":138,"logo":17,"time":77},1362799,"Deepmind's talent drain likely comes down to chip shortages, a conflict of interest, and Google's bureaucracy","https://the-decoder.com/deepminds-talent-drain-likely-comes-down-to-chip-shortages-a-conflict-of-interest-and-googles-bureaucracy/",{"id":140,"title":141,"source":142,"logo":16,"time":77},1351172,"Google seeks a sharper focus in AI after Hassabis move","https://www.ft.com/content/f860d189-7def-4c7a-8ce3-4e96b3371d1f?syn-25a6b1a6=1",{"id":144,"title":145,"source":146,"logo":28,"time":77},1351173,"Googlers Contend With Exit of AI Pioneer and ‘Most Google Person’","https://www.bloomberg.com/news/articles/2026-08-06/google-grapples-with-exit-of-ai-pioneer-and-most-google-person",{"id":148,"title":149,"source":150,"logo":35,"time":77},1355591,"Four of Google's Top AI Researchers Just Left to Start One Company. Alphabet Is Investing in It.","https://www.fool.com/investing/2026/08/06/four-of-googles-top-ai-researchers-just-left-to-st/",{"id":152,"title":153,"source":154,"logo":32,"time":72},1346987,"Exclusive / Demis Hassabis was shifting away from DeepMind CEO duties for a year","https://www.semafor.com/article/08/05/2026/demis-hassabis-was-shifting-away-from-deepmind-ceo-duties-for-a-year",{"id":156,"title":157,"source":158,"logo":31,"time":72},1345258,"Google DeepMind CEO Demis Hassabis is stepping aside","https://www.axios.com/2026/08/05/google-deepmind-demis-hassabis-ai",{"id":160,"title":161,"source":162,"logo":14,"time":72},1346990,"What Google’s AI Shake-Up Means for Pichai Succession","https://www.theinformation.com/newsletters/the-briefing/googles-ai-shake-means-pichai-succession",{"id":164,"title":165,"source":166,"logo":40,"time":72},1346991,"Big shake-up in Google’s AI team as DeepMind chief executive steps down","https://www.theguardian.com/technology/2026/aug/05/big-shake-up-in-googles-ai-team-as-deepmind-chief-executive-steps-down",{"id":168,"title":169,"source":170,"logo":5,"time":72},1345283,"Google’s AI Leadership Is Undergoing Some Major Changes. The Stock Is Sliding","https://www.investopedia.com/market-update-google-ai-leadership-is-undergoing-some-major-changes-the-stock-is-sliding-googl-12035042",{"id":172,"title":173,"source":174,"logo":30,"time":72},1355627,"Google’s AI shake-up: DeepMind’s Hassabis steps aside, senior scientists depart","https://arstechnica.com/gadgets/2026/08/googles-ai-shakeup-deepminds-hassabis-steps-aside-senior-scientists-depart/",{"id":176,"title":177,"source":178,"logo":22,"time":72},1355622,"Google Chief Scientist Who Helped Build Modern Search And AI Steps Away","https://www.searchenginejournal.com/google-loses-the-engineer-who-helped-build-search-and-modern-ai/584785/",{"id":180,"title":181,"source":182,"logo":25,"time":77},1349283,"What’s Behind the A.I. Shake-Up at Google","https://www.nytimes.com/2026/08/06/business/dealbook/hassabis-google-ai.html",{"id":184,"title":185,"source":186,"logo":19,"time":72},1355625,"Jeff Dean and other top AI researchers are leaving Google to launch their own startup","https://techcrunch.com/2026/08/05/jeff-dean-and-other-top-ai-researchers-are-leaving-google-to-launch-their-own-startup/",{"id":188,"title":189,"source":190,"logo":24,"time":72},1355621,"Google’s Mind Is Now Less Deep","https://spyglass.org/google-deepmind-hassabis-dean-gemini/",{"id":192,"title":193,"source":194,"logo":33,"time":77},1351180,"Google’s AI leadership comes apart in a single morning","https://www.fastcompany.com/91585546/googles-ai-leadership-comes-apart-in-a-single-morning",{"id":196,"title":197,"source":198,"logo":37,"time":199},1362801,"The next chapter of our AI momentum","https://blog.google/company-news/inside-google/message-ceo/next-chapter-ai-momentum/","3D AGO",{"id":201,"title":202,"source":203,"logo":5,"time":77},1351183,"Scientist Jeff Dean joined Google when it had less than 30 employees. 27 years later, he just quit to start over at 58","https://finance.yahoo.com/technology/ai/articles/scientist-jeff-dean-joined-google-173000017.html",{"id":205,"title":206,"source":207,"logo":21,"time":77},1362803,"Runtime: Bank of America's CTIO on AI costs and a big shakeup for Google AI","https://www.thestack.technology/runtime-bank-of-americas-ctio-on-ai-costs-and-a-big-shakeup-for-google-ai/",{"id":209,"title":210,"source":211,"logo":39,"time":77},1351181,"Google Builds A New Gemini Model Team As AI Investments Accelerate","https://www.nextplatform.com/cloud/2026/08/06/google-builds-a-new-gemini-model-team-as-ai-investments-accelerate/5284258",{"id":213,"title":214,"source":215,"logo":12,"time":77},1362804,"The Download: Google’s AI shake-up and Meta’s rogue model","https://www.technologyreview.com/2026/08/06/1141278/the-download-google-ai-shake-up-meta-rogue-model/","#6e6fbaff","#6e6fba4d",1786545065558]