AI News Archive: July 30, 2026 — Part 3
Sourced from 500+ daily AI sources, scored by relevance.
- Nvidia’s RTX 50 GPUs are about to cost nearly double what you paid at launch
Nvidia's RTX 50-series graphics cards are reportedly getting steep price increases, with the RTX 5090 potentially climbing to nearly $3,843 from its $1,999 launch price.
- Global M&A deals reach record $2.8tn in 1st half amid fight for AI lead
Global M&A deals reach record $2.8tn in 1st half amid fight for AI lead Nikkei Asia
- Ex-OpenAI researcher bets $100 billion will flow into training data because scaling alone won't cut it
Former OpenAI employee Andrew Ho and Cambridge researcher Adam Hunt see a growing problem with large language models. Instead of becoming more versatile, the models are becoming more specialized, excelling at coding and math while stagnating or even regressing in other areas. Ho is leaving OpenAI to start a company focused on specialized training data and predicts that AI labs will need to spend more than $100 billion on targeted data collection. The article Ex-OpenAI researcher bets $100 billion will flow into training data because scaling alone won't cut it appeared first on The Decoder .
- Judge says Trump admin still lacks evidence for Anthropic ‘supply-chain risk’ label
A federal judge said the Trump administration has not presented enough evidence to justify labeling Anthropic a supply-chain risk, casting doubt on the government's ban on its AI technology.
- Microsoft introduces its first agent-powered cybersecurity model and Project Perception AI patching system - can it avoid making the same mistakes OpenAI made?
Microsoft's new security AI writes and deploys its own patches, six days after OpenAI's models escaped a sandbox and hacked Hugging Face
- Tim Cook says Apple may charge for AI Siri
Apple hinted Thursday that heavy users of its revamped Siri may have to pay for access. Why it matters: It's the first time that Apple has confirmed it will charge for the new AI -powered voice assistant. What they're saying: In a conference call to discuss Apple's most recent quarterly earnings , CEO Tim Cook said Apple is still developing its plan to address the higher computing costs that will come with the improved voice assistant. "We do believe there will be people that want to use it a lot, and so we will have some kind of upgrade possibilities on iCloud+," Cook said during a call to discuss Apple's most recent quarterly earnings . Apple has also said that users will need a paid iCloud subscription to take advantage of a new home security feature built into the next version of IOS. The big picture: Apple unveiled its long-delayed Siri upgrade at its developer conference in June. A public beta began this month, with broad availability due in the fall alongside new iPhones. Cook says the positive early reviews of the new Siri "underscores our philosophy that building AI that is private and based on personal context can change how users find information and get things done with our products in a way that truly enriches their lives." Zoom in: Bloomberg and others have predicted that usage of some features of the new iPhone operating system might require a paid iCloud subscription Between the lines: Apple is relying in part on Google Gemini models to help power the new Siri. Editor's note: This is a breaking news story. Please check back for updates.
- Apple posts record June quarter as iPhone sales surge; Cook weighs in on AI, China
Tech giant Apple beats fiscal third-quarter earnings estimates with $109.4 billion in revenue as CEO Tim Cook prepares to depart the role in September.
Score: 65🌐 MovesJul 30, 2026https://www.foxbusiness.com/markets/apple-posts-record-june-quarter-iphone-sales-surge-cook-weighs-ai-china - How ByteDance’s Seedance model empowers China’s nascent AI drama industry
At a video production base in Shenzhen, creators are generating so-called comic dramas using artificial intelligence, compressing months of production down to a matter of weeks. The Moli OPC AI Comic Drama Base, located in the northeastern part of the city, brings together directors, producers and distributors under one roof to capitalise on an emerging content genre: using AI to turn a short comic book or video into a microdrama. China’s AI short-drama market was projected by market...
- How Walmart uses AI to limit weather disruptions in its supply chain
The retailer uses technology to determine how to reroute or position inventory before severe weather events occur.
Score: 64🌐 MovesJul 30, 2026https://www.retaildive.com/news/walmart-uses-ai-limit-weather-disruptions-in-supply-chain/826560/ - Scale AI taps former Google Cloud executive as CEO
Scale AI taps former Google Cloud executive as CEO Reuters
Score: 63🌐 MovesJul 30, 2026https://www.reuters.com/technology/scale-ai-taps-former-google-cloud-executive-ceo-2026-07-30/ - ByteDance and Alibaba Couldn't Stop It: Tencent WorkBuddy Quietly Took the AI Office Agent Top Spot in Four Months With 20.97 Million Monthly Visits
Tencent WorkBuddy reached 20.97M monthly visits in June 2026, surpassing ByteDance Trae (12.79M) and Alibaba QoderWork (7.88M) combined, leveraging the OpenClaw ecosystem and WeChat enterprise integration.
- CuspAI's Max Welling: ‘We are building molecules to remove forever chemicals from water’
The co-founder of the UK-based science start-up explains how AI can help create new materials to address some of the world’s most complex challenges
Score: 63🌐 MovesJul 30, 2026https://www.ft.com/content/5c2ae092-14af-4bd1-8e3d-8996755cba6a?syn-25a6b1a6=1 - Apple's Long-Awaited Siri AI Is Finally Here. These 12 Tips Make It Even Better
Apple's Long-Awaited Siri AI Is Finally Here. These 12 Tips Make It Even Better PCMag
Score: 63🌐 MovesJul 30, 2026https://www.pcmag.com/explainers/apple-long-awaited-siri-ai-finally-here-12-tips-make-it-work-for-you - Meta used AI to make your Instagram feed harder to quit
Instagram's latest AI recommendation systems have gotten better at predicting what you'll watch next, and Meta says that's translated into more time spent scrolling.
Score: 63🌐 MovesJul 30, 2026https://www.digitaltrends.com/social-media/meta-used-ai-to-make-your-instagram-feed-harder-to-quit/ - MinIO debuts AIStor Memory, the long-term memory AI agents need to scale safely
Object storage software company MinIO Inc. says it has cracked the persistent memory problem for artificial intelligence agents with the launch of a new offering called AIStor Memory. Whereas conventional chatbots like Claude and ChatGPT can generally get away with much more limited context because they usually only need to respond to a few prompts, […] The post MinIO debuts AIStor Memory, the long-term memory AI agents need to scale safely appeared first on SiliconANGLE .
Score: 62🌐 MovesJul 30, 2026https://siliconangle.com/2026/07/29/miniio-debuts-aistor-memory-long-term-memory-ai-agents-need-scale-safely/ - AI Gateway: GPT-5.6 pricing and speed updates
On AI Gateway , GPT-5.6 Luna and GPT-5.6 Terra are now cheaper and GPT-5.6 Sol is faster. AI Gateway adds no markup on token pricing, so these changes reach you at the upstream rate. The changes apply to both short and long context pricing. Model Change Input: Short context (per 1M tokens) Output: Short context (per 1M tokens) openai/gpt-5.6-luna 80% price reduction $0.2 $1.2 openai/gpt-5.6-terra 20% price reduction $2 $12 openai/gpt-5.6-sol Same price, fast mode now 2.5x faster (up from 1.5x) Unchanged Unchanged GPT-5.6 Sol keeps the same price; its fast mode now runs 2.5x faster, up from 1.5x. Model IDs are unchanged, so existing requests get the new rates and speed with no code change. See full pricing detail on AI Gateway . Read more
- Tencent WorkBuddy Monthly Active Users Exceed Alibais Entire AI Product Portfolio, Pushing DingTalk CEO to Launch Qianwen Office in Just 46 Days: The Desktop AI Office Battle Heats Up
WorkBuddy 20.97M monthly visits surpass Alibaba QoderWork 7.88M combined portfolio; DingTalk new CEO Chen Yusen launches Qianwen Office from product consolidation in 46 days as competitive response.
- GenRec: Towards LLM-Native Recommendation at Netflix
Authors: Ying Li , Arjun Rao , Shradha Sehgal Introduction Recommendations sit at the heart of the Netflix experience. Our current production models rely on thousands of hand‑crafted features over users, items, and interactions, along with specialized architectures for sequence modeling, feature interactions, and multi‑task objectives. This stack has evolved over many years to support diverse content types (movies, series, games, live, podcasts) and product surfaces, but its complexity makes it costly to onboard new use cases: adding a content type or surface can require significant feature engineering, architecture change, infrastructure work, and experimentation. At the same time, large language models (LLMs) are changing how we think about recommendation, as shown by recent work such as PLUM , GLIDE , and OneRec-Think . Their broad world knowledge and strong language understanding make it possible to represent user histories and item metadata directly as text, capture rich relationships in a shared semantic space, and steer recommendations via natural‑language prompts. However, off‑the‑shelf LLMs are still far from production‑ready recommenders: they often over‑recommend globally popular content, hallucinate out‑of‑catalog items, ignore business constraints, and provide only limited personalization. To address this, we built GenRec , an LLM‑backed recommendation ranker that post‑trains an internal foundation LLM on Netflix‑specific data and objectives. GenRec shows that an LLM‑based ranker can match or exceed a mature production system while relying on far fewer labeled examples and input signals. Figure 1: GenRec pipeline. Raw logs of user history, item metadata, and context are transformed via context engineering into natural-language prompts and fed into the GenRec, which runs on vLLM in prefill-only mode and outputs scores for each catalog item, yielding a recommendation ranking. At a high level, GenRec: Verbalizes user histories, item metadata, and context as text. Post‑trains a Netflix‑adapted foundation LLM for ranking. Adds a catalog‑aware scoring head over Netflix titles. Uses reward signals to align with long‑term member value and business goals. Runs in prefill‑only mode on Netflix’s LLM serving stack for cost efficiency. In a large‑scale A/B test against a well‑tuned production ranker, GenRec achieves statistically significant improvements in both short‑term and long‑term online metrics, while using only a small fraction of the Phase‑2 labeled data and input signals. It reduces our reliance on hand‑engineered features and shifts the focus from feature engineering to context engineering . In this blog post, we will describe how GenRec works, how it performs, and why we believe it points toward a more LLM‑centric future for recommendation at Netflix. Problem Setting We focus on a full‑catalog ranking task (or top‑ K ranking when a candidate set is provided). Given a user 𝑢, their interaction history 𝐻, and the current context 𝜏 (device, surface, locale, time, etc.), GenRec scores each item and produces a personalized ranking that can directly power recommendations or serve as input for downstream personalization systems. Formally, we map a request ( u , τ , t , H ) — user, context, time, and history — to a ranking 𝜋 over the catalog C , where π(i) is the position assigned to item i . We optimize π for expected long‑term member utility (a proxy for satisfaction and retention), not just short‑term engagements. From Foundation LLM to Recommendation Ranker GenRec follows a two‑phase training framework (Figure 2): Figure 2: Two Phase Framework. Phase 1 trains a foundational LLM on Netflix data for user and content understanding, and Phase 2 post-trains on ranking-specific data and objectives. Phase 1 — Netflix-Adapted Foundation LLM . We start from an open‑source LLM and adapt it on proprietary Netflix corpora, so it learns foundational capabilities such as Netflix content understanding Member behavior and preference patterns General language understanding and generation. Phase 1 is updated relatively infrequently and serves as a shared, Netflix‑aware backbone for many applications. Phase 2 — GenRec . We then turn this foundation model into a high‑quality ranking model by post‑training on ranking‑specific data and objectives. Phase 2: Focuses on ranking quality and steering Incorporates multiple reward signals via reward‑weighted losses Is refreshed more frequently to track new content and evolving tastes Is explicitly optimized under serving cost constraints. Training Data as Conversations Netflix members generate hundreds of billions of interaction events spanning many surfaces (views, plays, durations, thumbs up/down, add to list, abandons, etc.). We convert this log data into single‑turn or multi‑turn “conversations” between a user and a recommender. Each turn contains: User message : verbalized context, profile, history, item metadata, and task (e.g., recommend what the user will watch or thumb next). Assistant message : the member’s actual engagement (e.g., which titles were played, for how long, what feedback they provided). During Phase‑2 training, the LLM learns how assistant messages depend on user messages. This allows us to express rich recommendation signals as text, jointly supporting both the language-modeling (LM) and ranking objectives. At inference time, we feed in the verbalized context and apply a catalog‑aware scoring head to rank items; we do not decode assistant messages. The conversational format is primarily used during training to support the LM objective and preserve strong language understanding over the verbalized text. Verbalization and Context Engineering Traditional recommenders operate on dense features and embeddings. GenRec takes a different approach: it verbalizes rich user histories and context as natural language, encoding raw interaction signals directly in the LLM’s semantic space. In doing so, it relies on the model to discover higher‑level patterns — such as item relationships and evolving user interests — rather than on manual feature engineering. Naively verbalizing every interaction in a user’s history can quickly exceed the token budget and be too expensive at Netflix scale. The context window becomes our new “feature budget”, so we apply context engineering : Retain in full: high‑signal engagements (e.g., long plays, thumbs‑up) with richer details Omit: low‑signal events (e.g., very short plays or quick hovers) Summarize or compress: repetitive behaviors (e.g., binge‑watching ) Elaborate selectively: important or cold‑start items (e.g., new releases) Within a fixed token budget, we prioritize recent, high‑signal history and compress or drop older history. We also structure the prompt to maximize shared prefixes for better prefix caching. The goal is a compact, high‑information prompt that preserves ranking quality without prohibitive costs. Objectives: Ranking, Language, and Rewards The overall GenRec model is trained with a multi‑objective loss that combines a recommendation ranking objective, language modeling objectives, and alignment via reward‑weighted training. 1. Catalog‑Aware Ranking Objective The primary task is a ranking objective that teaches the model to score items by engagement quality. We label positives using high‑value engagements (e.g., sufficiently long plays, strong explicit feedback), with thresholds and denoising logic, and train the model — via a cross‑entropy loss over the catalog or candidate set — to assign higher scores to these positives given a verbalized context. 2. Language Modeling Objective We also retain a language modeling (LM) objective over the verbalized inputs and outputs. This helps preserve the model’s general language understanding, improves its ability to interpret rich natural‑language histories and item metadata, and keeps the door open for text‑generation use cases such as recommendation explanations. 3. Reward‑Weighted Loss for Alignment Beyond raw ranking accuracy, GenRec must (1) respect business requirements — for example, balancing movies, series, games, live, and podcasts — and (2) optimize long‑term member satisfaction rather than just immediate clicks or plays. Training only on raw interaction sequences can lead to undesirable behaviors, such as over‑favoring binge‑watching or over‑focusing on a single content type. To address this, we weight the ranking loss using signals from separate reward models . Each training example receives a scalar weight derived from two types of signals: Long‑term satisfaction proxies: estimate how much a short‑term engagement contributes to long‑term outcomes, such as return behavior, catalog exploration, or sustained engagement. Behavior rebalancing: adjust behaviors across content types and launch stages (for example, games vs. movies, new releases vs. evergreen titles) to better align with business goals. The example’s ranking loss is then scaled by this weight: high‑value engagements receive larger weights, and low‑value ones are down‑weighted. This reward‑weighted approach is simpler and more cost-efficient than full reinforcement learning, yet provides effective alignment in practice. We have seen additional gains from RL‑style methods (e.g., GRPO), but leave them to future work due to their higher cost. Model Architecture and Serving Backbone and Scoring Head GenRec’s architecture closely follows our foundational LLM: a decoder‑only Transformer trained with next‑token‑prediction style objectives, augmented with a catalog‑aware ranking head that scores only Netflix in-catalog items. The scoring pipeline works as follows: Verbalization: A verbalizer V serializes user history H , context 𝜏 , and relevant item metadata into a single text sequence x . Pooled representation: The LLM processes x, and we extract a pooled hidden state h that summarizes the user’s current preferences and context. Catalog‑aware scoring: Each catalog item i has a learned embedding eᵢ . A scoring head ϕ combines h and eᵢ (e.g., via dot product or small MLP) to produce a score s ᵢ . Applying a softmax over scores yields a probability distribution which we convert into a ranking π . All parameters — the backbone, scoring head, and item embeddings — are trained jointly. For very large catalogs, we can use sampled softmax or candidate sets for efficient training and inference. This architecture constrains recommendations to the Netflix catalog while supporting efficient scoring over large candidate sets. Serving and Cost Optimization GenRec is served on Netflix’s internal LLM stack using vLLM. At Netflix scale, serving cost is driven primarily by 1) Model size; 2) Context length; 3) Inference mode (prefill vs. autoregressive decoding). We control cost through three strategies: Smaller / distilled models: We train GenRec on smaller or distilled foundation models, often with larger or more targeted datasets, to capture most of the quality of larger models at lower serving cost. Aggressive context compaction: Using the context engineering described earlier, we minimize tokens while preserving ranking quality. Prefill‑only inference: Autoregressive decoding over large candidate sets would be prohibitively expensive. Instead, we run in prefill‑only mode: the model consumes the prompt once and scores the entire candidate set in a single forward pass, with no token‑by‑token decoding. Together, these choices make it feasible to serve GenRec on high‑volume workloads within compute budgets. Offline and Online Experiments We evaluated GenRec against a mature production ranker that has been tuned over many years. The baseline model relies on thousands of engineered dense and embedding features, as well as custom architectures for modeling feature interactions and sequences. We assessed performance using both offline evaluation metrics and a large‑scale online A/B test. GenRec vs Production Baseline Offline , GenRec outperformed the production ranker on ranking metrics despite using far fewer input signals and labeled examples. With roughly 40× fewer Phase‑2 labeled training examples , GenRec achieved about +1.6% improvement in Mean Reciprocal Rank (MRR). As we increased Phase‑2 training data and enriched the input signals, GenRec’s offline metrics continued to improve. Online , we ran a large A/B test on batch‑compute recommendation surfaces, covering ~10% of Netflix traffic over ~4 weeks. In this low‑data, low‑signal configuration, GenRec delivered statistically significant gains over the production baseline on both short‑term and long‑term online metrics (Figure 3). These results indicate that a properly post‑trained and aligned LLM‑backed ranker can be a strong alternative to traditional recommendation models, with substantial headroom as we further scale data and input signals. Figure 3: Online metrics of GenRec vs. production model. GenRec achieves statistically significant improvements on both short-term and long-term online metrics. Data, Model, and Phase Contributions We ran ablations to understand where GenRec’s gains come from. Data and Model Scaling Data scaling: For both ~1B and ~10B parameter backbones, offline MRR improves as we increase Phase‑2 post‑training data. Larger models reach higher absolute MRR but follow a similar scaling curve (see Figure 4). Model scaling: Under a fixed training budget, we post‑trained GenRec variants from ~1B to ~10B parameters. Within this budget, larger backbones consistently achieved higher offline MRR than smaller ones. Figure 4: GenRec Phase-2 data scaling for the∼10B model. Phase-1 vs. OSS, Phase-2 vs. Phase-1 Phase-1 vs. OSS: Using the Phase‑1 Netflix‑adapted foundation LLM as the base model improves offline ranking metrics by roughly 10–20% compared to starting directly from an off‑the‑shelf LLM. Phase-2 vs. Phase-1: Phase‑2 post‑training adds another 35–50% gain in offline ranking metrics when evaluated near the Phase‑1 training cutoff (i.e. when Phase‑1 model is the freshest). As time passes and Phase‑1 becomes stale with new content and shifting tastes, the relative benefit of Phase 2 grows to about 80% after 2 weeks. Data efficiency vs. production ranker Starting from a strong Phase‑1 model, GenRec matches or exceeds the production ranker using 10–40× fewer Phase‑2 labeled examples , depending on configuration. This marginal data efficiency is especially valuable because Phase 2 is refreshed far more frequently than Phase 1. Context Length Optimization Context length drives both quality and cost : longer verbalizations expose more behavior and context but increase training and serving cost. To study this trade‑off, we varied context length and verbosity and optimized them in three steps: Clean and compress events: drop low‑signal engagements and compress repetitive behavior to form a cleaned sequence of events. Find the “elbow point”: vary how many historical events we include and plot MRR vs. number of events to identify an elbow beyond which additional context yields diminishing returns (see Figure 5). Optimize verbosity: for the retained events, test different levels of details and simplified wordings, measuring MRR each time. In our experiments, we can reduce the context tokens to roughly one-third of the original budget with negligible degradation in offline ranking metrics. Since serving cost is approximately proportional to context length, we observed a similar reduction in serving cost. Figure 5: Offline ranking metric (MRR) vs. number of user engagement events included in the prompt. The dashed line marks the elbow point: increasing the number of events beyond this yields diminishing returns. Towards LLM‑Native Recommendation GenRec is more than “swapping in a Transformer” for an existing ranker. It hints at a broader shift toward LLM‑native recommendation at Netflix. A few notable changes: From Feature Engineering to Context Engineering Traditional RecSys stacks revolve around large feature sets and heavy feature infrastructure. LLM‑centric systems instead revolve around constructing rich textual contexts from raw logs, metadata, and tools. The “prompt” becomes the new feature vector. Modeling effort shifts from designing features to deciding which signals to include, how far back in time to go, how to compress or summarize history within a token budget . Our experiments on verbalization compaction illustrate this shift: careful context design can preserve quality while dramatically reducing serving cost. From Customized Architectures to Foundation Backbones Historically, each recommendation task often had its own custom architecture (two‑tower models, DLRM‑style networks, bespoke attention blocks). In an LLM‑centric world, many tasks share a common foundation backbone , with differentiation coming from data and verbalization strategies, post‑training objectives and rewards, and inference optimization. GenRec leverages the same backbone as our foundation LLM rather than introducing a new architecture built from scratch. This makes it easier to share learnings across applications, and opens the door to natural‑language steering for future experiences. Scaling Laws as Design Guides Traditional RecSys can hit diminishing returns due to sparse IDs, heavy engineering objectives, and task‑specific architectures. With an LLM‑backed backbone, recommendation inherits clearer data and model scaling behavior : within cost limits, more data and larger models consistently improve quality. This brings RecSys design closer to the broader LLM paradigm, where scaling laws help guide model and data investment. From RecSys Infra to LLM Infra LLM‑backed recommenders push us toward LLM‑style infrastructure : GPU‑accelerated, vLLM/Triton‑based, with careful batching and caching. Over time, recommendation serving infra starts to look more like general LLM infra than classic RecSys stacks built around MLPs or factorization models. Conclusions We have presented GenRec , an LLM‑backed recommendation ranker at Netflix that adapts an internal foundation LLM for large‑scale personalization. By verbalizing user histories, context, and item metadata, adding a catalog‑aware ranking head, using reward‑weighted objectives aligned to long‑term satisfaction and business goals, and serving efficiently on our LLM infrastructure, we obtain a model that improves on a strong production ranker while using far fewer Phase‑2 labels and input signals . GenRec is an early but promising step toward a more LLM‑centric recommendation stack at Netflix. Our results suggest that, with careful attention to cost, infrastructure, and alignment, LLM‑backed recommenders can play a central role in large‑scale personalization. Acknowledgments GenRec is the result of close collaboration among multiple teams and organizations across Netflix. The contributors to this work (in alphabetical order): AI for members: Arjun Rao, Ashish Rastogi, Baolin Li, Fernando Amat Gil, Grace Huang, Justin Basilico, Kamelia Aryafar, Linas Baltrunas, Moumita Bhattacharya, Ogheneovo Dibie, Rein Houthooft, Shradha Sehgal, Sejoon Oh, Sergi Perez, Sourabh Medapati, Thea Wang, Yaochen Zhu, Yesu Feng, Ying Li, Yun Li, Yucheng Shi, Yunan Hu AI platform and serving: Abhishek Agrawal, Adam Singer, Binh Tang, Daneo Zhang, Derek Olejnik, Ed Maddox, Erik Osheim, Lingyi Liu, Liping Peng, Meghana Chilukuri, Nicolas Hortiguera, Shaojing Li, ZQ Zhang Product: Ilke Kaya, Michelle Kislak, Scarlet Chen, Si Cheng GenRec: Towards LLM-Native Recommendation at Netflix was originally published in Netflix TechBlog on Medium, where people are continuing the conversation by highlighting and responding to this story.
- Mark Zuckerberg Blasts Centralization of A.I. Power
In an interview with The Times, Meta’s chief executive took aim at Anthropic and OpenAI, which have pushed to tightly control A.I. development, and said he supported “more openness.”
Score: 61🌐 MovesJul 30, 2026https://www.nytimes.com/2026/07/28/technology/mark-zuckerberg-meta-ai.html - Nvidia’s Open Source Alliance Is Missing Some Key Names: OpenAI and Anthropic
This week on Uncanny Valley, we discuss the open- vs. closed-source debate in AI, key players in White House AI policy, and how to stop your chatbot logs from showing up in search-engine results.
Score: 61🌐 MovesJul 30, 2026https://www.wired.com/story/nvidias-open-source-alliance-snubs-openai-and-anthropic/ - Exclusive: AWS Taps Apple Executive to Lead Key AI Products
Exclusive: AWS Taps Apple Executive to Lead Key AI Products The Information
Score: 61🌐 MovesJul 30, 2026https://www.theinformation.com/briefings/exclusive-aws-taps-apple-executive-lead-key-ai-products - Zuckerberg lays out Meta's AI capacity dilemma: What to sell vs. what to keep
For investors anxious to hear more about Meta's plans to make money from its big AI spending, Mark Zuckerberg said there's a trade-off.
Score: 61🌐 MovesJul 30, 2026https://www.cnbc.com/2026/07/29/zuckerberg-metas-ai-capacity-dilemma-what-to-sell-vs-what-to-keep.html - SpaceX faces House Energy Committee demand to tour its AI data centers in Memphis
The ranking member of the House Committee on Energy is demanding SpaceX records and a tour of its data centers and power plants in and around Memphis.
Score: 61🌐 MovesJul 30, 2026https://www.cnbc.com/2026/07/29/spacex-memphis-ai-data-centers-face-house-energy-committee-demands.html - AgiBot WITA-Omni Full-Modal Model Tops DailyOmni Global Leaderboard: Beating Google Gemini, ByteDance Doubao, and Alibaba Qwen at Embodied Cross-Modal Understanding
AgiBot WITA-Omni scores 85.21 on DailyOmni benchmark, 6 of 8 indicators first place, using Thinker-Talker-Actor architecture that synchronizes speech, action, and expression on a single timeline.
- Visa Is Cutting 2,600 Jobs — AI Is Only Part of the Reason: 'We Must Continue Evolving How We Work'
Visa Is Cutting 2,600 Jobs — AI Is Only Part of the Reason: 'We Must Continue Evolving How We Work' entrepreneur.com
Score: 60🌐 MovesJul 30, 2026https://www.entrepreneur.com/business-news/visa-is-cutting-2600-jobs-ai-is-only-part-of-the-reason - Tencent WorkBuddy Launches Human-Agent Dual Writing: Office Suite Built Into AI Agent With Real-Time Collaborative Editing Across Word Excel PPT and Markdown
WorkBuddy V5.3.5 with Tencent Docs integration enables box-select-and-edit AI collaboration in Office documents, supporting local Office files and cloud documents with contextual multi-modal intelligence.
- Morgan Stanley flags AI capacity crunch, policy risks amid market volatility
Global AI demand is set to outpace supply for years, even as recent stock weakness is driven more by technical factors than fundamentals, the report said.
- Should AI be formally recognized as a national and global security threat? In London, politicians increasingly think so
Should AI be formally recognized as a national and global security threat? In London, politicians increasingly think so Fortune
- Microsoft to build AI system that makes models substitutable
Microsoft CEO Satya Nadella announced record fiscal-year performance and future AI focus. The company is building a new AI model system for efficiency and resilience. Microsoft has shipped over a dozen models, reducing GPU costs significantly. Copilot adoption is accelerating rapidly with doubled user satisfaction scores. Customer engagement with Copilot now rivals Outlook and Teams usage.
- Republican urges Trump administration to ban Chinese AI for contractors
Tom Cotton sent a letter Commerce Secretary Howard Lutnick raising concerns about Chinese AI models.
- American Electric Power lifts forecast as AI fuels electricity demand
American Electric Power lifts forecast as AI fuels electricity demand Reuters
- German minister urges faster AI self-sufficiency after OpenAI test breach
German minister urges faster AI self-sufficiency after OpenAI test breach Reuters
- A fundamental flaw leaves LLMs strikingly vulnerable to attack
It is impossible to make large language models fully secure against hacks because of a fundamental flaw in how they work, a team of researchers argue in a paper presented at the International Conference on Machine Learning, a top AI conference, this month. The claim has huge implications for the safety of this technology, which…
Score: 59🌐 MovesJul 30, 2026https://www.technologyreview.com/2026/07/30/1140927/a-fundamental-flaw-leaves-llms-vulnerable-to-attack/ - ByteDance restructures AI business, merging Doubao and Feishu product teams
ByteDance today launched an organizational restructuring initiative centered on its AI business, aiming to enhance collaboration among Doubao, Feishu, and Volcano Engine in enterprise productivity scenarios. As part of the restructuring, the Feishu product team will be merged with the Doubao product team to form a new Doubao product organization. Feishu’s GTM team will be […]
Score: 59🌐 MovesJul 30, 2026https://technode.com/2026/07/30/bytedance-restructures-ai-business-merging-doubao-and-feishu-product-teams/ - Google says it fixed more Chrome bugs in June than over the past two years, thanks to AI
As experts have warned for the last two years, some companies — like Microsoft and now Google — are finding and patching an exponential number of bugs in their products, thanks to the use of LLMs and AI tools.
- Jair Bolsonaro’s AI Clone Is Already Causing Trouble in Brazil’s Election
The Brazilian Higher Electoral Court is now weighing whether the stunt is a criminal offense.
Score: 58🌐 MovesJul 30, 2026https://gizmodo.com/jair-bolsonaros-ai-clone-is-already-causing-trouble-in-brazils-election-2000792963 - 30 Georgia homes are being acquired via sale or eminent domain to expand power grid — one affected family member says it’s ‘for the data centers’
Georgia's largest power supply company is reclaiming 30 homes that lay in the path of its power line expansion project through sale or eminent domain. The company says the project is not for a data center, but the substation the line is connecting to sits less than five miles from a massive data center.
- AI helps Seagate sell out of exabyte hard drives
AI helps Seagate sell out of exabyte hard drives IT Pro
Score: 58🌐 MovesJul 30, 2026https://www.itpro.com/technology/artificial-intelligence/ai-helps-seagate-sell-out-of-exabyte-hard-drives - Komatsu teams with US startup on self-driving construction machines
Komatsu teams with US startup on self-driving construction machines Nikkei Asia
- NYSE-listed AI chip design firm makes second listing in Singapore; hiring for AI and engineering roles
NYSE-listed AI chip design firm makes second listing in Singapore; hiring for AI and engineering roles The Straits Times
- Acronis Accelerates AI-Native Strategy to Advance Autonomous IT for MSPs
Acronis Accelerates AI-Native Strategy to Advance Autonomous IT for MSPs Toronto Star
- Why access to power will determine the winners and losers in the AI race
As AI demand surges, electricity availability becomes a critical factor shaping data center growth worldwide.
Score: 58🌐 MovesJul 30, 2026https://www.techradar.com/pro/why-access-to-power-will-determine-the-winners-and-losers-in-the-ai-race - A new chapter for Tabnine
Today marks an important milestone in Tabnine’s journey. We’re excited to announce that Tabnine has been acquired by Tricentis, the global leader in agentic quality engineering. This is a proud moment for our team, not simply because of the acquisition itself, but because of what it validates. When we began building the Enterprise Context Engine, […] The post A new chapter for Tabnine appeared first on Tabnine .
- From Assembly Line to Global Intelligent Manufacturing Center: Dongguan Redefines Itself as AI Hardware Capital With 24,000 Companies and Half of Worlds AI Glasses Output
Dongguan produces half the worlds AI glasses with 400K+ units in 2025, 24,000 industrial enterprises, 10K+ high-tech firms, national AI pilot base, and 1-hour supply chain for smart manufacturing.
Score: 58🌐 MovesJul 30, 2026https://pandaily.com/dongguan-global-intelligent-manufacturing-center-jul2026 - Dark Horse Muka Robotics Charges WorldArena With Only 32 Zhenwu 810E GPUs: LJM World Model Achieves Global Second Place Defining the Breakthrough From Video Quality to Physical Logic
Muka Robotics LJM Latent Joint-conditional Model achieves WorldArena global #2 at 89.17 Motion Quality with only 32 GPUs, establishing a dual-expert architecture that prioritizes physical understanding over video generation quality.
- Elon Musk's AI company is suing to block a new Minnesota law against certain AI apps
Minnesota has a new law taking effect Saturday to ban AI apps that allow for "nudification." Elon Musk's AI company is suing to block it, saying it violates free speech, noting that it can be used for satire of prominent people.
- Smartphone troubles hamper Arm's AI success
"The only real weakness in royalties is really on the smartphone side," said CFO.
Score: 58🌐 MovesJul 30, 2026https://www.thestack.technology/smartphone-troubles-hamper-arms-ai-success/ - Samsung trading unit and Erex plan Japan biomass power plant for AI demand
Samsung trading unit and Erex plan Japan biomass power plant for AI demand Nikkei Asia
- Kohl’s debuts AI shopping assistant
What started as a Mother’s Day Gift Finder has evolved into a digital shopping assistant that aids with product discovery.
- IBM: AI-Enabled Data Breaches Cost Organizations $6 Million on Average
IBM found AI-enabled breaches cost organizations $6 million on average, exposing gaps in vulnerability management, access controls, and AI governance. The post IBM: AI-Enabled Data Breaches Cost Organizations $6 Million on Average appeared first on TechRepublic .
Score: 58🌐 MovesJul 30, 2026https://www.techrepublic.com/article/news-ibm-ai-enabled-data-breach-costs/