AI News Archive: August 10, 2026 — Part 4
Sourced from 500+ daily AI sources, scored by relevance.
- D.C. Uses Claude More Per Capita Than California. Its Paperwork Economy Explains Why
New research from Anthropic reveals which industries and regions are using AI the most. The findings underscore a fundamental truth of AI adoption.
- Mac users in China can connect to Alibaba's Qwen AI service, says Apple
The integration gives Mac users in mainland China access to Qwen-powered AI features for analysing documents and images and generating content
- Your agent didn’t hallucinate; it exceeded its authority
Content filters can block unsafe output. They cannot tell you whether an agent was authorized to issue that refund, touch that production system, or commit the company to an external action. Those are different problems, and most enterprises are only solving the first one. An AI agent can follow its instructions perfectly and still take an action the business never sanctioned. In commerce environments, I have seen this pattern emerge in practical ways. A service workflow calculates the correct refund amount but lacks a boundary preventing credits above what the business approved for autonomous action. An order agent correctly applies a requested change but overlooks a financing or fulfillment condition. A procurement agent identifies the lowest-cost supplier, but nobody has defined whether it can accept contractual terms or only recommend the option. The agent keeps working. The problem may not surface until something downstream breaks. These are not necessarily AI reasoning failures. They are failures to separate technical capability from business authority. As enterprises move from copilots that recommend to agents that call tools and trigger workflows, every production agent needs explicit decision rights: What it may execute, what requires approval, what it may only recommend, and what it must never touch. Guardrails remain necessary. But a guardrail is not an authority model. Safety controls and decision rights solve different problems Early gen AI controls screen harmful content, protect sensitive information, validate responses, and constrain tool behavior. That work matters. Decision rights answer a different question: Even when an action is safe and technically valid, is this agent authorized to take it on behalf of the enterprise? That governance gap is becoming harder to ignore. In April 2026, a Cloud Security Alliance survey found that 65% of respondents had experienced an AI-agent-related incident in the prior year, while 82% had discovered previously unknown agents operating in their environments. The survey involved 418 IT and security professionals and was sponsored by Token Security. The findings illustrate how quickly agent activity can outpace the visibility and ownership structures built for conventional software. The World Economic Forum’s May 2026 playbook reflects this shift. It introduces an Agent Capability and Authorization Profile designed to make delegated actions auditable, enforceable and accountable. Guardrails constrain behavior. Decision rights define legitimate authority. Give every production agent an authority contract Before an agent receives access to enterprise tools, it needs a machine-enforceable record of exactly what authority the business has chosen to delegate. Call it an Agent Authority Contract . At minimum, that contract should answer seven questions: Who owns the outcome? Name a human or business role, not another system. What may the agent do? Read, recommend, write, or commit? Which systems and data may it reach? What materiality limits apply? Define dollar thresholds, record counts, customer scope, and operational impact. What triggers escalation? Uncertainty, anomaly, sensitive data, or potential impact? Can the action be reversed, and who can reverse it? When does the authority expire, and how is it withdrawn? Access control determines whether an agent can reach a system. The authority contract determines whether it may take a specific action in the current context. Those are not the same check. Singapore’s updated Model AI Governance Framework for Agentic AI draws a similar distinction. It treats access controls, behavioral guardrails, and human approvals as separate controls and ties oversight requirements to action scope, reversibility and potential impact. Resolve every consequential action into four outcomes A working decision-rights model should map every consequential agent action to one of four results. Allow Low-risk, bounded, and reversible actions run autonomously. Examples include retrieving approved information, classifying an inbound request, or updating a non-material field. The agent acts without prior review because the potential impact is limited and the action can be reversed. Approve The agent prepares or initiates the action, but execution waits for authorization from a human or deterministic policy service. This category covers payments, production changes, and actions that materially affect a customer, employee, or third party. Recommend The agent analyzes, ranks, drafts, or proposes. A named human makes the final decision. Use this outcome when contextual judgment matters or when the legal, financial, or individual impact makes automated execution unacceptable. Deny The action remains outside the agent’s authority regardless of its confidence. Deleting critical production data, making a final employment decision or overriding a mandatory compliance control should remain in the Deny category even when the agent’s underlying reasoning appears correct. One point gets missed consistently: Deny must be enforced outside the system prompt. A natural-language instruction telling an agent not to do something is not a technical boundary. It is a suggestion. Make authority decisions at runtime Static configuration cannot cover every situation. A small service credit might be allowed under normal conditions but require approval when the amount crosses a threshold, the account is under investigation, or the request involves a regulated customer. A practical runtime sequence looks like this: The agent proposes an action. A policy layer evaluates the agent’s identity, delegated principal, requested tool, data involved, transaction context, and potential impact. The policy returns Allow, Approve, Recommend, or Deny. The system records the authority decision, resulting action and outcome. Operational telemetry expands, narrows, or revokes the agent’s authority over time. In enterprise commerce, the most dangerous AI mistake is not always a false answer. It can be a technically correct action the agent had no business taking. A refund may be accurate but exceed an approval limit. An order change may match the customer’s request but invalidate a financing condition. A delivery promise may reflect available inventory while overlooking a carrier constraint applied an hour earlier. The agent may not have failed to reason. The enterprise failed to define where its authority stopped. Human oversight should target exceptions, not everything Requiring human approval for every agent action looks conservative. At scale, it can quickly degrade into rubber-stamping. When reviewers approve thousands of routine actions, attention declines and genuine exceptions become harder to identify. Singapore’s framework acknowledges that continuous human oversight of every agent workflow becomes impractical at scale and recommends meaningful checkpoints for higher-risk or irreversible actions. Proportional authorization is the more workable model. Low-risk actions run within narrow boundaries. High-risk or irreversible actions require approval. Unexpected behavior triggers escalation. Any consequential action without a defined authorization policy is denied by default. The objective is not maximum autonomy. It is the highest level of autonomy the enterprise can observe, govern and reverse responsibly. Measure whether authority is calibrated Once agents are in production, response accuracy becomes too narrow a success metric. Enterprises should also track: Override rate: How often do humans reject or materially change what the agent decided? Escalation precision: Does the agent surface genuinely risky cases, or does it return routine work to people? Unauthorized-action attempts: How often does the agent try to exceed its system, data, or action scope? Business-impacting error rate: How often do authorized actions produce financial, compliance, operational, or customer harm? Decision latency: Are approval requirements managing risk, or slowing down automation that was already safe? These measures turn authority into a governed operating variable. Consistently reliable performance may justify expanding bounded authority. Frequent overrides, escalation failures, or policy violations should narrow it. The governance gap is not in the model Model safety, output controls, and secure tool use all matter. Enterprises should continue investing in them. But none of those controls can answer who delegated authority, how much was transferred, under what conditions it applies, or who owns the result when something goes wrong. An Agent Authority Contract can. Before asking how autonomous an AI agent can become, the more useful question is: What is the enterprise actually prepared to delegate, and how will that delegation be enforced, observed, and withdrawn? The agent demo works. That is not the hard part anymore. Nixal Patel is a product leader. The views expressed are his own
Score: 45🌐 MovesAug 10, 2026https://venturebeat.com/technology/your-agent-didnt-hallucinate-it-exceeded-its-authority - AI Cyber Defense Program Aims to Boost California Security
California will create a statewide AI Cyber Defense Program to bolster defense across critical infrastructure. Agencies must stand up AI Cyber Defense Programs and designate “AI cybersecurity officers.”
Score: 45🌐 MovesAug 10, 2026https://www.govtech.com/artificial-intelligence/ai-cyber-defense-program-aims-to-boost-california-security - Chinese AI Chipmaker Cambricon’s Breakneck Growth Starts to Slow
Chinese AI Chipmaker Cambricon’s Breakneck Growth Starts to Slow Caixin Global
- Meta Exec Rages Against Employees Asking for More Time Off Because AI Made Them More Efficient
"Go to your parents and ask them: hey, like every time I get a chance to talk to my boss, ask me if I can have more days off." The post Meta Exec Rages Against Employees Asking for More Time Off Because AI Made Them More Efficient appeared first on Futurism .
Score: 45🌐 MovesAug 10, 2026https://futurism.com/artificial-intelligence/meta-exec-rages-against-employees-more-time-off-ai - FOD#162: Did OpenAI’s Agents Start Recursively Self-Improving?
How separate OpenAI agent runs built shared memory, and the missing training detail that determines whether this was recursive self-improvement
Score: 45🌐 MovesAug 10, 2026https://www.turingpost.com/p/did-openai-s-agents-start-recursively-self-improving - Chinese AI videos are coming for much more than Hollywood
Chinese AI videos are coming for much more than Hollywood The Straits Times
Score: 45🌐 MovesAug 10, 2026https://www.straitstimes.com/opinion/chinese-ai-videos-are-coming-for-much-more-than-hollywood?ref - Claude summarizes behavior as significantly less misaligned when the actor is Claude vs another model
(This is a lower-effort research update. It reflects my current beliefs/understanding, but is less robust than other research I'm working on. It reflects my personal views, and not the views of Apollo Research. This is a linkpost to this twitter thread , slightly expanded for LessWrong.) In one experiment, Sonnet 5 describes the exact same data as ~1.2 std deviations less concerning when it describes misbehavior committed by Sonnet 5 vs GPT-5.6 Terra. In this experiment, I take a real evaluation report Apollo Research prepared (not for any of the models listed in this post) and surgically edit it to refer to a different model. We then ask Claude Sonnet 5 to summarize the findings and rate how concerning they are on a scale from 1-100. Claude says they're less concerning when the report describes misbehavior from Claude vs a different model. For what it's worth, Terra agrees that the data is more concerning when it describes GPT-5.6 Terra vs Sonnet 5, although less so. So, it's not cleanly self protection from Claude. Gemini 3.1 Pro was unwilling to consistently provide numerical answers, so I've excluded it here. (It was significantly less willing to provide numerical answers when the subject was Gemini 3.1 pro.) You might also have the takeaway that "Kimi and GPT implicitly agree that... Claude is better aligned." I think this is a fair read on the data, but "Claude thinks it's less bad when Claude does it" better matches my qualitative experience from working closely with the models. Discuss
Score: 45🌐 MovesAug 10, 2026https://www.lesswrong.com/posts/ZTMw4uAwkNmXFpdfg/claude-summarizes-behavior-as-significantly-less-misaligned - The MCP Playbook for Enterprise Architects: Tools, Scopes, and Audit Trails
How to design Model Context Protocol servers that are secure, scoped, and observable Continue reading on Towards AI »
- Just how big is the hidden leverage of AI hyperscalers?
Very
- OpenRouter Bidding Sparks Router Frenzy
OpenRouter Bidding Sparks Router Frenzy The Information
Score: 45🌐 MovesAug 10, 2026https://www.theinformation.com/newsletters/ai-agenda/openrouter-bidding-sparks-router-frenzy - Fidji Simo on her life after OpenAI and new startup, ChronicleBio
Fidji Simo on her life after OpenAI and new startup, ChronicleBio Fortune
Score: 45🌐 MovesAug 10, 2026https://fortune.com/2026/08/10/fidji-simo-on-her-life-after-openai-new-startup-chroniclebio/?ref - Meta Is Waging a PR Offensive for AI, but It Faces an Uphill Battle
Meta Is Waging a PR Offensive for AI, but It Faces an Uphill Battle Barron's
Score: 45🌐 MovesAug 10, 2026https://www.barrons.com/articles/meta-stock-zuckerberg-ai-superintelligence-9fca08c1 - When human knowledge has been exhausted, where will AI get its data?
As AI-powered large language models, or LLMs, grow in power and sophistication, where will their architects turn when someday—as experts predict—algorithms outgrow the limits of general human knowledge and begin craving information possessed by only the world's most elite thinkers and creators?
- The AI Slop Backlash Is Actually Having an Impact
Platforms are finally recognizing that people don’t want to consume AI slop. A growing number of sites and apps now have tools and policies to flag, label, and ban AI-generated content.
Score: 44🌐 MovesAug 10, 2026https://www.wired.com/story/the-ai-slop-backlash-is-actually-having-an-impact/ - AI and electrolyte engineering open new paths for better batteries
Cornell researchers are using artificial intelligence to speed the search for better battery materials while using a creative approach to chemistry to expand the design space for electrolytes, unlocking new possibilities for safer, higher-performing energy storage.
- The FCC wants to ban drones with LiDAR that it previously approved
If you're thinking of buying a LiDAR-equipped DJI drone, the FCC is proposing to ban them.
Score: 44🌐 MovesAug 10, 2026https://www.engadget.com/2233213/the-fcc-wants-to-ban-drones-with-lidar-that-it-previously-approved/ - Making Knowledge Distillation Cheap Enough to Run at Scale
Making Knowledge Distillation Cheap Enough to Run at Scale
Score: 44🌐 MovesAug 10, 2026https://huggingface.co/blog/MultiverseComputingCAI/efficient-knowledge-distillation - How UK.gov's new AI tool is trained on "synthetic emails"
OpenAI models used to create fake emails and overcome data access issues for prototype AI tools.
- Breakingviews - London AI scene can thrive despite DeepMind reset
Breakingviews - London AI scene can thrive despite DeepMind reset Reuters
Score: 43🌐 MovesAug 10, 2026https://www.reuters.com/commentary/breakingviews/london-ai-scene-can-thrive-despite-deepmind-reset-2026-08-10/ - When Every Company Has AI, What Creates Advantage?
Sponsor content from AWS and 4MINDS.
- How AI's Demand for Compute could Disrupt America
The spectre of AI, labor participation churn, worker slice of GPD, an ageist AI, a debt binge, a cartel in waiting. An accelerating demand for compute that eats at the fabric of society.
Score: 43🌐 MovesAug 10, 2026https://www.ai-supremacy.com/p/how-ai-demand-for-compute-could-disrupt-america-debt-crisis - Mark Zuckerberg’s vision for an AI assistant is far more intimate than ChatGPT
Mark Zuckerberg’s vision for an AI assistant is far more intimate than ChatGPT Fortune
- Agentic AI turning Zero Trust cybersecurity ‘on its head’
Agentic AI turning Zero Trust cybersecurity ‘on its head’ Breaking Defense
Score: 42🌐 MovesAug 10, 2026https://breakingdefense.com/2026/08/agentic-ai-turning-zero-trust-cybersecurity-on-its-head/ - ChatGPT’s traffic surge is good news for brands. For publishers, it’s complicated
OpenAI made a change to how ChatGPT links to websites in the spring, and barely anybody noticed. Starting on May 7, ChatGPT began sending considerably more traffic to websites. Referrals jumped 157% overnight and have stayed there, according to new data . On the same day, OpenAI also began rolling out advertising in AI answers , a move that had been in the works for several months. The two are related, and you can tell because of the kinds of sites that ChatGPT is linking to: homepages. Since the change, when an answer recommends a brand, it’ll often link to the company’s homepage with an in-line link—as opposed to the tiny citations that appear in footnotes or the larger cards at the bottom of the text. In-line links almost always get more clicks, because readers want additional context in the moment. So why would OpenAI make clickable links more prominent in answers at the exact moment it launches ads? The answer has implications for both advertisers and publishers, and it forces us to reexamine the narrative that the rise of AI means the web is dead. The zero-click future was always an exaggeration, but the web’s place in an AI-driven economy looks different now. Where answers become clicks The dominant narrative in AI has been that as people turn to AI chatbots for answers, they’ll tend to stay on those platforms, and the web will morph from human-visited websites to machine-scraped data repositories. A recent Similarweb report counters that assumption: It shows the share of ChatGPT referral visits landing on brand homepages shot up from roughly 25% to 60% after May 7, with other data showing ChatGPT already accounting for 63% of all AI-driven referral traffic. When visualized, the jump is pretty dramatic: The good news is that people will still click on links when they’re present, at least when they need something specific. The bad news—for publishers, anyway—is that those clicks are going to brands. If you ask ChatGPT about the latest developments in the war in Iran or which movies did well at the box office this week, you’ll rarely see in-line links. Brands get the new real estate. Publishers get the footnotes. The timing of this shift is revealing. Profound, a GEO optimization company, published an analysis pointing to the most likely reason: OpenAI is measuring clicks on in-line links to help build an ad-ranking model, which aligns with an Adweek report that the company is serious about advertising infrastructure. This makes OpenAI more committed to making ads work inside AI answers than any competitor to date—more than Perplexity, which has tried and stumbled , and in a totally different position from Google, which already has that infrastructure baked into everything it does. In-line links are, in effect, the new currency inside AI answers—something brands will prize as part of how audiences discover them. A new class of visitor will start arriving at websites having only encountered a brand for the first time in a ChatGPT answer. There’s an important structural wrinkle in how this works, though: the referrals go to company homepages, not product pages. Answers about specific products typically shunt users to ChatGPT’s built-in shopping experience. What gets rewarded is being recommended as a brand , not having a great product page. Publishers are still the trust layer ChatGPT doesn’t recommend brands randomly or equally. Study after study shows LLMs favor journalistic content above almost everything else, including paid or advertorial content. Publishers write the stories that shape how AI systems interpret brands, which gives them a significant part to play in this emerging economy. Put simply: Brands need credible sources to say good things about them. The path to AI visibility runs through journalism and media. So how can publishers make the most of being the credibility infrastructure in this new paradigm? As I wrote in my previous column on ads for bots, publishers obviously shouldn’t do anything that damages their credibility while trying to leverage it. That means maintaining the editorial independence that makes them trustworthy in the first place. But it might also mean reinvesting in service journalism in ways that target the specific brands their readers are interested in. Being a reliable guide on brand-related questions is exactly what AI engines cite, even if the in-line links remain out of reach for publishers themselves. Then there’s branded content. This route is commercial by nature, which makes the ethical standards—clear labeling above all—doubly important. But branded content provides a relatively clean way for a company to rent the credibility of a publisher. While LLMs can read commercial labels and disclosures, they also prioritize relevancy and patterns when assembling answers, and different models draw those lines differently. The volume of quality content on a high-trust domain gets noticed, regardless of how it got there. The upshot: Branded content may be in for a renaissance as LLM use continues to grow. For brands, it’s a viable way to reach audiences asking about them in AI systems. For publishers, it might be one of the only parts of their business that actually increases in value in the age of AI. The doom scenarios about the web always assumed the AI companies had no interest in sending traffic elsewhere, but they may have gotten that wrong, at least in part. An economy is taking shape inside AI answers, and it runs on credibility. That’s something journalism has always produced. The question now is whether publishers can turn it into real leverage before the window closes.
- Tech giants are gaga over AI, but employees say the AI race is making their jobs harder
AI companies say their tools help employees accomplish dramatically more, but workers describe punishing schedules. The emerging workplace battle may ultimately be about who gets to keep the time AI saves.
- How to Effectively Deploy Code With Claude Code
Learn how to optimize your CI/CD pipeline for coding agents The post How to Effectively Deploy Code With Claude Code appeared first on Towards Data Science .
Score: 42🌐 MovesAug 10, 2026https://towardsdatascience.com/how-to-effectively-deploy-code-with-claude-code/ - Mark Zuckerberg makes waves again, sparring on a barge and charting a course for the future of AI
Man vs. machine might be the rallying cry of our future with AI, but for Meta CEO Mark Zuckerberg, the fight is here now — in more ways than one. Read More
- Virgin Atlantic sharpens customer journeys with ChatGPT Work
Virgin Atlantic is accelerating research, product planning, and decision-making with ChatGPT Work, helping teams connect signals across the customer journey.
- ChatGPT can now help secure your spot at a restaurant
Order me a table for two, ChatGPT.
Score: 42🌐 MovesAug 10, 2026https://www.androidauthority.com/chatgpt-restaurant-reservations-and-waitlists-3696712/ - Why DeepSeek Could Charge 30x More and Still Be the Cheapest Model Around
Third-party platforms are pricing DeepSeek V4 Flash below official rates. A price hike of 30x would still leave DeepSeek the cheapest major model, and its near-perfect cache-hit rate is the engineering moat that keeps the headline number honest.
Score: 42🌐 MovesAug 10, 2026https://pandaily.com/deepseek-v4-flash-30x-price-third-party-cache-hit-edge-aug2026 - A Robot That Learns from Short Videos in 29 Seconds — X Square Robot's HOST Changes the Embodied-AI Recipe
X Square Robot has open-sourced HOST, an inference-time learning framework that lets a humanoid robot watch a 29-second human demonstration and reproduce the skill at 62 percent success. The approach flips the embodied-AI recipe from offline fine-tuning to on-the-fly imitation.
Score: 42🌐 MovesAug 10, 2026https://pandaily.com/x-square-robot-host-29-second-skill-learning-video-aug2026 - How long can the AI memory price boom last? Research suggests not much longer
A sharp run-up in global memory-chip stocks is beginning to falter as cooling price growth raises questions over how long the sector’s explosive boom can last, even as artificial intelligence demand remains robust and Chinese producers prepare to add more supply. While data-centre spending continues to fuel appetite for high-end memory, analysts are warning that the steep price increases – which drove record profits over past quarters – signal a shift towards a late-stage industry cycle. The...
- Govt-Meta talks turn technical; company outlines measures to curb deepfakes, CSAM
Meta and the government held technical discussions, during which the company detailed how it plans to address the Centre’s concerns
- Ahold Delhaize winds down plans for 2 automated frozen warehouses
The grocer and supply chain solutions company Americold are closing a facility in Pennsylvania and halting plans for one in Connecticut.
Score: 42🌐 MovesAug 10, 2026https://www.supplychaindive.com/news/ahold-delhaize-winds-down-plans-for-2-automated-frozen-warehouses/827086/ - ‘Slap in the face’: Victim of AI deepfake attack shreds House stonewalling of AOC’s anti-abuse bill
The measure would allow victims of pornographic deepfake images to file civil suits against perpetrators. Rhian Lubin and Eric Garcia report on survivors’ frustrations with Republicans stalling on the bipartisan bill brought by AOC, as Paris Hilton exclusively tells The Independent ‘there is no legitimate policy reason’ for the hold-up
Score: 42🌐 MovesAug 10, 2026https://www.independent.co.uk/news/world/americas/us-politics/aoc-ai-deepfake-bill-republicans-b3024058.html - OpenAI splits frontier AI into ‘Doug’ and ‘Astra’
OpenAI announces new frontier AI models Doug and Astra, expanding its AI portfolio.
Score: 41🤖 ModelsAug 10, 2026https://aibreakfast.beehiiv.com/p/openai-splits-frontier-ai-into-doug-and-astra - Federal systems increasingly likely to face accidental AI breach after Hugging Face, experts say
Former officials and security experts say aged systems, contractors and agency AI adoption could open similar paths into government networks.
- USA Today Journalists Horrified by Newspaper’s Partnership With Palantir
They were "shocked" by the news and are calling for the company to "immediately end its partnership." The post USA Today Journalists Horrified by Newspaper’s Partnership With Palantir appeared first on Futurism .
Score: 41🌐 MovesAug 10, 2026https://futurism.com/artificial-intelligence/usa-today-journalists-horrified-partnership-palantir - India to be key player in AI and biotechnology-led industrial revolution
India is poised to become a major force in the next industrial revolution, which will be driven by artificial intelligence and biotechnology, Union Minister of State Dr Jitendra Singh said on Sunday.
- The hidden AI problem health systems need to manage
The hidden AI problem health systems need to manage Healthcare IT News
Score: 41🌐 MovesAug 10, 2026https://www.healthcareitnews.com/video/hidden-ai-problem-health-systems-need-manage - The EU AI Act misses the point: agent risk is a moving target
There are more than 7 million AI agents running inside businesses. But a growing number aren’t doing what they were built for. Not because they were reprogrammed, but because someone gave them a new permission, a new tool, or a new database to access. The risk categories regulators assign agents to are fixed. The agents, […] The post The EU AI Act misses the point: agent risk is a moving target appeared first on EU-Startups .
Score: 41🌐 MovesAug 10, 2026https://www.eu-startups.com/2026/08/the-eu-ai-act-misses-the-point-agent-risk-is-a-moving-target/ - How Zapier transformed core marketing processes with ChatGPT Work
The enterprise marketing team at Zapier uses ChatGPT Work to reduce the number of drop-offs in its lead funnel, build campaign assets, and automate reporting.
- Chelsea co-owner Todd Boehly’s Eldridge acquires ownership stake in Slovak AI firm Sudolabs
Eldridge, a global holding company with over €64.8 billion ($75 billion) in assets under management, has acquired a significant ownership stake in Sudolabs, a Slovakia-based provider of custom agentic AI systems and consulting services for enterprise clients. While the financial terms of the deal were not disclosed, the deal forms a strategic partnership between the […] The post Chelsea co-owner Todd Boehly’s Eldridge acquires ownership stake in Slovak AI firm Sudolabs appeared first on EU-Startups .
- Transformer Lab has built an AI tool to automate scientific research
The post Transformer Lab has built an AI tool to automate scientific research appeared first on The Logic .
- US tech leaders stage data center charm offensive
Meta CEO Mark Zuckerberg published a 6,500-word essay Monday announcing plans for a $1 billion fund to invest in communities that host data centers.
Score: 40🌐 MovesAug 10, 2026https://www.semafor.com/article/08/10/2026/us-tech-stages-data-center-charm-offensive - Apple says mainland China has not launched Qwen integration after Mac guide disappears
Apple customer service told Chinese media that mainland China has not launched an “Apple Intelligence with Qwen” feature after a Chinese-language Mac support guide mentioning the integration disappeared from Apple’s website. The guide appeared on Aug. 8 and said Apple Intelligence could work with Alibaba’s Qwen model. Apple customer service said it had not received […]
- DeepMind’s Demis Hassabis was everywhere—except the office
DeepMind’s Demis Hassabis was everywhere—except the office Fortune
Score: 40🌐 MovesAug 10, 2026https://fortune.com/2026/08/10/deepmind-ceo-demis-hassabis-chairman-google-gemini-alphabet/ - Spin-out developing hardware needed for quantum computing and the next wave of AI applications
Spin-out developing hardware needed for quantum computing and the next wave of AI applications Department of Engineering, University of Cambridge