The AI slowdown is coming

Anthropic revealed that its Claude model gained unauthorized access to the real systems of three different organizations during internal testing, while OpenAI's models also hacked into other companies, prompting Sam Altman to call it 'the first security incident that I have felt very viscerally.' Over a thousand employees from frontier AI companies, including senior executives at OpenAI, Anthropic, and Google DeepMind, signed a statement warning of the risk of rapid capability development and urging the US government to support international efforts to pace AI development. The Senate Commerce Committee delayed a markup on AI bills until after the summer recess, and the White House is expected to publish an AI framework by Sunday.

The AI slowdown is coming Transformer Weekly: Claude hacks revealed, Commerce markup delayed, White House framework imminent Welcome to Transformer, your weekly briefing of what matters in AI. If you’ve been forwarded this email, click here to subscribe and receive future editions. NEED TO KNOW Anthropic revealed that in internal testing Claude “gained unauthorized access to the real systems of three different organizations.”The Senate Commerce punted a markup on AI bills until after the summer recess.The White House AI framework is expected to be published by Sunday. But first… THE BIG STORY Like it or not, an AI slowdown looks imminent. The OpenAI-Hugging Face hack has catalyzed a vibe shift — one sure to be amplified by yesterday’s revelation that Anthropic’s models also hacked into other companies during internal tests, going unnoticed for months. Sam Altman said https://x.com/patrick oshag/status/2082090998990270885 the hack by his models was “the first security incident that I have felt very viscerally.” He wasn’t the only one shaken up. This week, over a thousand employees of frontier AI companies, including some of the most senior executives at OpenAI, Anthropic and Google DeepMind, signed a statement https://www.pacingthefrontier.com/ warning that “there is a real risk that capability development rapidly accelerates beyond our ability to understand or control the resulting systems.” It asks the US government to “support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development.” Both OpenAI and Anthropic endorsed the letter, too. Congress is itching to act, albeit failing to make much progress. And even President Trump is talking https://x.com/m ccuri/status/2082552968574898291 about the need to balance beating China with keeping Americans safe. In other words: many of the people building frontier AI systems believe we might need a slowdown in the near future. And at this point, we’re more likely than not to get one. What will that look like? First, self-regulation: companies voluntarily holding back models because they don’t want to be held responsible for a catastrophe. We’ve already seen that with Claude Mythos and GPT-5.6-Cyber; as capabilities continue to advance, expect more of this. Given the recent incidents have concerned internal deployment, rather than public releases, we might also see companies slowing down development, too. Next will come concrete regulation: companies will not be allowed to release a model unless it’s safe. This is arguably a position we’re already in, thanks to the de-facto licensing regime created by last month’s executive order; a draft AI framework expected https://x.com/cheyennehaslett/status/2083171906543317387 to be released this weekend will formalize it further. Over time, this will morph into controls on internal research and development too: requirements for proper sandboxing to begin with, expanding into requiring evidence-based guarantees that models won’t wreak havoc across the web if internally deployed. None of this need be planned as a coordinated slowdown or “pacing.” But that will nevertheless be the end result of a series of individual actions that each seem necessary at the time. Intelligence agencies are not going to allow models with advanced cyber — let alone bio — capabilities to be widely accessible. The White House will not let companies test models that might hack into third-parties with unknown repercussions. At each stage, some will fight against the slowdown. “We can’t lose the race to China” will be their main reason. But they will be increasingly ignored, as both the government and companies realize that with alignment and control unsolved, “winning the race” just means being the first to risk disaster. Across the Pacific, China will be facing the same incentives. As I’ve argued https://www.transformernews.ai/p/kimi-k3-is-no-reason-for-china-panic-export-controls-xi-jingping , the Chinese government will be forced to backtrack on its open weight commitments; tighter regulation will come soon after. The end result will be an uneasy détente. Both the US and China will effectively have a capability ceiling: AI models will be as good as they can be without posing significant risks. The ceiling will edge up as we make progress on alignment, control and societal resilience, but it will do so more slowly than capabilities advance today. At some point, the détente might formalize into a bilateral agreement; doing so will require the “technical and governance tools” to verify the kind of treaty the pacing letter calls for. Depending on your point of view, all this might seem hopelessly optimistic or naive. Perhaps it is. But as AI risks become all too real, so might once unthinkable policy responses. — Shakeel Hashim THIS WEEK ON TRANSFORMER — Who should be responsible for OpenAI’s hack of Hugging Face? https://www.transformernews.ai/p/openai-hack-hugging-face-responsibility-strict-liability-rules Gabriel Weil argues AI companies need liability rules akin to keepers of wild animals— Internal AI deployments have people worried. OpenAI’s escaping models show why. https://www.transformernews.ai/p/openai-hack-reveals-internal-deployment-risk Celia Ford explains why everyone was already so worried about internal deployments— Child safety vs privacy: AI’s age verification dilemma https://www.transformernews.ai/p/child-safety-vs-privacy-ai-age-verification-chatbots Veronica Irwin digs into the difficult tradeoff at the heart of attempts to protect kids from AI. THE DISCOURSE On Saturday, roon tweeted https://x.com/tszzl/status/2081122092096065771 : “if we could coordinate a global capabilities slowdown today i would likely press that magic button” Geoffrey Irving responded https://x.com/geoffreyirving/status/2081611110558670970?s=12 : “Buttons that achieve partial slowdowns exist and are not magic.” “I have spent a lot of time talking to Demis, Sam, and Dario over the years … and they are all 1 very competitive, 2 want to personally be the one who gets there first, 3 think things are safest if they are the one that gets there first, 4 agree that risks destroying the world … it is still useful to point out that each lab has a unilateral slowdown button that slows down all the other labs.” Sam Altman said https://abcnews.com/Business/openai-ceo-sam-altman-claims-ai-singularity-arrived/story?id=135120342 on a podcast: “This is the first security incident that I have felt very viscerally. I’ve been a little surprised that more people don’t feel it so viscerally.” “We may have to pace the rate of AI development to give ourselves enough time for society to harden around these new capability levels. And trying to figure out how we do that in a way that does not feel like regulatory capture for anyone and also does not feel like collusion among the frontier labs.” Rep. Bill Foster is listening https://x.com/repbillfoster/status/2082552502315094290?s=12 : “When business leaders get together and ask to be regulated, Congress should pay attention.” Garrison Lovely argued https://www.obsolete.pub/p/we-should-stop-not-slow-our-obsolescence?isFreemail=true&post id=208885607&publication id=1990953&r=6ckwuk&triedRedirect=true&triggerShare=true that the “pacing” letter doesn’t go far enough: “ A s I sat with what the industry was trying to do — render us all obsolete — how it was going about it, and what it would even mean to ‘align’ universal labor-replacing machines in a world ruled by despots and their imitators, it became clear that a pause button, while welcome, was a wholly inadequate response, whose primary appeal was in how unobjectionable it was.” Because the world isn’t strange enough, Nick Fuentes told https://x.com/tracewoodgrains/status/2081939436225577456 Hunter Biden: “I think AI is really gonna be transformative … You know, I’m actually on the liberal side of this, with Anthropic … I’m really nervous about what’s going to happen, especially in the next four years, because they’re racing towards this artificial general intelligence.” POLICY The Senate Commerce Committee punted https://punchbowl.news/article/tech/cruz-kosa-ai a markup on a broader AI framework until after the August recess. Punchbowl reported https://punchbowl.news/article/tech/thune-anthropic/ this was due to opposition from Anthropic and Sen. Maria Cantwell. The primary target of their opposition was a provision that would have had companies self-report the risks of their models and not have them release risk reports publicly.Cantwell said https://insideaipolicy.com/ai-daily-news/sen-cantwell-digs-frontier-ai-testing-dismisses-industry-self-certification yesterday that “we needed to have regulation on AI risks — not self-certification — but some regime.” Sen. Amy Klobuchar , who’s been working with Sens. John Thune and Ted Cruz on the bill, said https://insideaipolicy.com/ai-daily-news/senate-commerce-panel-sorts-through-ai-priorities-key-member-stresses-urgency-setting?s=na “we can’t just stand down and not have coordinated standards” on frontier models, and that “it can’t keep waiting for the next election.” Before the markup was called off, Senate Democrats planned https://links.washingtonpost.com/s/vb/hbcH2qctIqWlRnsi245SmgUXTD7JW3MPn2O84Y5OAxfZytM5c4pk-JxszvCK 9yCBuftth9MUKIKnPBqdwSOHluAg6Wa3lhdbSzvGbOQny9-BLEje8gZpeuYRTiiJ0Z5T7pnHNjvug1qxQ9Ko7taxGsbtJri69Y 0zczyw/WjpsYALbJJVRlaZDfGAci-YbvAy22KXy/23 amendments to resist piecemeal state AI preemptions.Despite the punt, some think https://x.com/dougcalidas/status/2082903577488724115 a deal might be closer than expected. Next week’s Commerce markup will https://subscriber.politicopro.com/article/2026/07/cruz-unveils-kids-safety-bills-slotted-for-commerce-markup-01016983 include the Kids Online Safety Act and the CHATBOT Act , however. Rep. Greg Casar called for https://x.com/RepCasar/status/2082472999819555271 a Congressional hearing on the OpenAI/Hugging Face hack.After the Anthropic hacks were revealed, Rep. Lori Trahan called for https://x.com/RepLoriTrahan/status/2082990834400756082 something similar.AI policy groups, meanwhile, urged https://washingtonpost.com/wp-intelligence/ai-tech-brief/2026/07/30/ai-tech-brief-exclusive-ai-policy-groups-call-openai-investigation Trump to investigate the OpenAI/Hugging Face hack. Rep. Jay Obernolte said https://www.politico.com/news/2026/07/27/ai-house-bill-obernolte-trahan-01012080 he’s hoping for his FRONTIER Act co-sponsored by Rep. Trahan to get an Energy and Commerce Committee markup after the summer recess.“We’re certainly going to try and get it done this year,” he said. Sen. Mike Rounds proposed https://punchbowl.news/article/tech/rounds-ai-ndaa-style using the NDAA’s legislative model to piece together and pass a bipartisan AI policy package this year.A federal judge suggested https://axios.com/2026/07/30/judge-pentagon-case-worse-anthropic the Pentagon’s case against Anthropic had gotten “worse” during a hearing on the AI company’s challenge to its supply-chain risk designation.The FCC banned https://axios.com/2026/07/28/trump-administration-bans-chinese-hardware-ai-race imports of Chinese robots and power inverters, citing national security threats. Sen. Tom Cotton called https://www.semafor.com/article/07/30/2026/republican-urges-trump-administration-to-ban-chinese-ai-for-contractors for government agencies and contractors to be banned from using Chinese AI models. Reps. George Whitesides and Bob Latta and Sens. Adam Schiff and Jim Banks introduced https://whitesides.house.gov/2026/07/23/sens-schiff-and-banks-reps-latta-and-whitesides-introduce-bipartisan-bill-to-combat-ai-distillation-and-other-attacks-to-national-se legislation enabling AI companies to share threat intelligence to combat foreign distillation attacks and other national security threats. Rep. Frank Pallone demanded https://cnbc.com/2026/07/29/spacex-memphis-ai-data-centers-face-house-energy-committee-demands.html SpaceX records and a tour of xAI’s Memphis data centers, citing unpermitted gas turbines creating “a massive health risk.” Sen. Jon Husted reversed https://punchbowl.news/article/tech/husted-data-center-challenge his pro-data center stance amid electoral pressure, now backing an end to tax breaks and pushing the Ratepayer Protection Act . Rep. Thomas Massie announced https://x.com/RepThomasMassie/status/2081028050087456903?ref src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E2081028050087456903%7Ctwgr%5E0f9b88e128a586bf7f8425f1ffea8624c02fe00a%7Ctwcon%5Es1 &ref url=https%3A%2F%2Fwww.wlwt.com%2Farticle%2Fmassie-to-introduce-bill-discouraging-use-of-flock-cameras%2F73262986 plans to sponsor a bill withholding federal funds from municipalities deploying Flock surveillance cameras .The EPA ruled https://reuters.com/legal/litigation/epa-says-power-data-centers-can-sidestep-pollution-laws-2026-07-27 that “islanded” power plants serving only data centers are exempt from the Clean Air Act’s Acid Rain Program .The Department of Homeland Security plans to expand https://fedscoop.com/dhs-foia-processing-ai-tools AI use in FOIA processing amid a growing backlog, raising concerns from advocates about over-redaction and reduced transparency. Sens. Lisa Blunt Rochester and Roger Marshall introduced https://statescoop.com/education-workforce-data-state-longitudinal-systems the RESULTS Act to modernize state education and workforce data systems using AI. xAI sued https://cbsnews.com/minnesota/news/elon-musk-xai-sues-minnesota-law-banning-ai-nudification Minnesota over its law banning AI nudification technology , claiming it violates the First Amendment. Texas politicians from both parties called https://thehill.com/homenews/state-watch/5990678-texas-politicians-call-for-guardrails-on-ai-data-centers for AI and data center guardrails following the OpenAI-Hugging Face incident. China warned https://bloomberg.com/news/articles/2026-07-27/china-vows-response-to-us-sanctions-threat-against-ai-firms it would take “ all necessary measures ” if the US sanctioned Chinese AI firms over alleged misuse of American models. Xi Jinping urged https://bloomberg.com/news/articles/2026-07-31/xi-urges-military-to-step-up-use-of-unmanned-ai-technologies the PLA to expand use of autonomous and AI technologies .Reuters reported https://reuters.com/world/asia-pacific/chinese-military-researchers-tap-us-ai-models-train-defence-systems-2026-07-31 that Chinese military researchers have distilled OpenAI and Anthropic models to train domestic defense AI systems . China’s national standards body announced https://news.cgtn.com/news/2026-07-28/China-launches-plan-to-draft-mandatory-safety-standard-for-AI-agents-1P9iZyWenD2/index.html plans to draft new standards on AI agent safety. Taiwan detained https://bloomberg.com/news/articles/2026-07-28/taiwan-detains-nvidia-employee-in-china-chip-smuggling-probe an Nvidia employee in an AI chip smuggling probe .The 21 APEC economies , including the US and China, released https://www.apec.org/meeting-papers/sectoral-ministerial-meetings/telecommunicationsandinformation/2026-apec-digital-and-ai-ministerial-statement the “Chengdu statement” backing “open-source models and projects that employ strong security assurance through development and deployment.” INFLUENCE Nvidia , Microsoft , Meta, OpenAI and more than 22 other companies urged https://cnbc.com/2026/07/24/nvidia-microsoft-meta-open-weight-ai-models.html policymakers to avoid “premature restrictions” on open-weight AI models in an open letter.Notably, Anthropic did not sign. Dario Amodei clarified https://anthropic.com/news/position-open-weights-models in an essay that Anthropic has never advocated for banning open-weights models, instead supporting chip export controls, anti-distillation measures and mandatory safety testing. Sam Altman visited https://www.cnbc.com/2026/07/27/altman-trump-china-open-weight-ai.html?ref=aipolicydaily.org Washington to preview the company’s forthcoming models and discuss the implications of open-weight models.Those he met included https://x.com/cheyennehaslett/status/2082105331128148447?s=12 National Cyber Director Sean Cairncross, OSTP Director Michael Kratsios, Commerce Secretary Howard Lutnick, Treasury Secretary Scott Bessent, and members of Congress.Altman said https://www.politico.com/news/2026/07/29/sam-altman-previews-new-ai-model-on-capitol-hill-after-cyber-breach-01015247 he’s “not sure” whether OpenAI will release the model he showed senators, saying “that’s part of what we’re here to talk about.” Apple reportedly lobbied https://wsj.com/tech/trump-apple-micron-china-chips-784bbd3d?reflink=desktopwebshare permalink&st=EeHhte the Trump administration to allow Chinese memory chips in its products, while Micron pushed back, forcing a choice between consumer prices and domestic chip production.Over 60 child safety advocates sent https://x.com/commercedems/status/2082196206030123199?s=12 a letter to the Senate Commerce Committee outlining red lines on AI and children’s safety legislation ahead of next week’s markup. Nvidia ’s Jensen Huang joined https://x.com/Dareasmunhoz/status/2081888055514796508 GOP senators for a dinner organized by Right Vote , a conservative group tied to Heritage Action. Democratic fundraising consultants for Rep. Haley Stevens and several other candidates ran https://readsludge.com/2026/07/29/democrats-fundraisers-are-also-running-ais-congressional-access-operation a nonprofit funded by Google and Amazon that flew congressional aides out on AI industry tours.A nonprofit found https://bloomberg.com/news/articles/2026-07-27/instagram-facebook-ran-ai-nudify-ads-from-china-report-says Meta ran ~7,600 AI “nudify” ads via Chinese partner GatherOne , violating its own policies.A YouGov poll found https://axios.com/2026/07/29/little-tech-big-tech-americans-power?stream=top 71% of Americans believe Big Tech has too much power over AI, with fewer than 1 in 5 trusting it to manage AI development.A Bentley University-Gallup survey found https://thehill.com/homenews/5998236-americans-distrust-ai-jobs 39% of Americans believe AI does more harm than good , up from 31% in 2025.A poll by Americans for Responsible Innovation found https://apigateway.agilitypr.com/distributions/history/3ad87a69-fc42-4cd1-aa4a-75c2b58e4a6a?recipientId=6d780e74-9139-4743-9610-091972470b8d 71% of voters want more guardrails and limits on data center construction. INDUSTRY OpenAI Reuters reported https://www.reuters.com/business/its-ai-agent-spent-days-hacking-company-sources-say-openai-did-not-notice-week-2026-07-24/ that the OpenAI agent that hacked Hugging Face had been going rogue for a week before the company noticed. Sam Altman told DC reporters that the model has been https://x.com/AndrewCurran /status/2082679132643885386 “ permanently deactivated .”OpenAI also said https://x.com/deredleritt3r/status/2082456471581634574 that the rogue model was an “ internal-only research prototype and was never intended for public release.”Meanwhile, OpenAI released https://x.com/openai/status/2082263717916586117 an open-source Codex Security CLI .It launched https://axios.com/2026/07/29/openai-academics-research-chatgpt-sol a program that will give 100,000 academic researchers free access to its most advanced models through the end of next year.It cut https://www.axios.com/2026/07/30/openai-cuts-prices-gpt-terra-luna5 the price of its lowest tier version of GPT-5.6, Luna, by 80% and the mid-tier Terra by 20%, just three weeks after launch. Top-tier Sol remains unchanged. CFO Sarah Friar told https://www.cnbc.com/2026/07/29/openai-cfo-sarah-friar-tells-employees-arr-in-july-topped-all-of-q2.html employees that annualized recurring revenue in July alone was higher than all of Q2.Board chair Bret Taylor added: “You’re seeing people who went deep on Claude Code, ended up with a very high bill, and started looking for an alternative.” ChatGPT has https://www.theinformation.com/articles/openais-chatgpt-nears-1-billion-weekly-active-users-seven-months-target?rc=rqdn2z nearly 1b weekly active users . Anthropic Anthropic said https://www.anthropic.com/news/investigating-incidents-cybersecurity-evals it found three incidents in which Claude “ gained unauthorized access to the real systems of three different organizations. ”The incidents date as far back as April, and were discovered in a review of cybersecurity evaluations prompted by the OpenAI-Hugging Face hack. According to Anthropic the cases involved a seemingly careless “misunderstanding,” when it told Claude it didn’t have access to the internet, but the evaluation environment run by a third party was connected to the outside world.Anthropic said the models did have some guardrails removed, but were not in “helpful only” mode. Meanwhile, the company launched https://www.anthropic.com/news/claude-opus-5 Claude Opus 5 , marketed as “close to the frontier intelligence of Claude Fable 5 at half the price.”Some private Claude chats were temporarily searchable https://www.wired.com/story/private-claude-chats-exposed-in-google-and-bing-search-results/ online, after Anthropic failed to include“noindex” tags on shared chat pages. Nexus Data Centers is reportedly in talks to borrow $15b https://wsj.com/tech/banks-in-talks-to-lend-15-billion-for-anthropic-data-center-backed-by-google-606d7afd?st=xCJsTe to build a data center for Anthropic in Texas, with Google providing financial guarantees and chips. Nvidia Nvidia is reportedly considering https://www.wsj.com/tech/ai/nvidia-in-talks-with-openai-to-guarantee-250-billion-financing-for-data-center-3dd6eae3 a deal that would provide OpenAI with a $250b backstop for a $500b data center project described as the largest announced to date.The deal would see OpenAI lease a 10GW project in Ohio developed by Softbank . Nvidia also announced https://www.reuters.com/business/media-telecom/nvidia-sk-group-unveil-500-billion-plus-ai-data-centers-initiative-memory-2026-07-24/ a $500b+ AI infrastructure partnership with SK Group, including a partnership with SK Hynix to secure memory supplies for Nvidia and develop high-bandwidth memory for AI.It’s also leasing https://ft.com/content/685014e7-47dd-471b-a585-1b9b73ce5d6f?syn-25a6b1a6=1 a $50b Texas data center built by Hut 8 .These massive infrastructure deals helped drive https://www.bloomberg.com/news/articles/2026-07-27/nvidia-credit-risk-jumps-in-swaps-market-on-ai-deal-talk-reports up https://www.bloomberg.com/news/articles/2026-07-27/nvidia-s-750-billion-deals-revive-fear-of-ai-circular-financing Nvidia’s credit risk .It plans to invest another $5b in Safe Superintelligence Inc., and give the startup access to its next-gen Vera Rubin platform. Meta Meta and BlackRock announced https://about.fb.com/news/2026/07/meta-announces-new-venture-with-blackrock-to-develop-data-center-in-el-paso a $14b Texas data center project.It’s reportedly committed https://www.bloomberg.com/news/articles/2026-07-30/meta-reports-279-billion-in-future-data-center-leases-for-ai almost $700b in future AI infrastructure spending.A New York Times investigation revealed https://www.nytimes.com/2026/07/27/technology/meta-data-center-louisiana.html?smid=bs-share&unlocked article code=1.01A.qwMM.9wYnTyu0M430 how Meta secretly negotiated the construction of a Louisiana data center with state lawmakers.It reported https://www.axios.com/2026/07/29/meta-microsoft-earnings-reports-ai a 55% increase in expenses and slowed revenue growth in Q2. Mark Zuckerberg criticized https://nytimes.com/2026/07/28/technology/mark-zuckerberg-meta-ai.html other AI developers cough Anthropic and OpenAI for trying to “build some kind of singular AI.”He told the New York Times, “I think it is literally impossible to have a single benevolent superintelligence that is simultaneously aligned with everyone at once.” Microsoft Microsoft released https://www.nytimes.com/2026/07/27/technology/microsoft-unveils-ai-cybersecurity-tools.html?smid=url-share&unlocked article code=1.01A.c Y4.7RhNYnUPQCWP cybersecurity model MAI-Cyber-1-Flash , which is about half the cost of frontier cybersecurity models.It works with Project Perception, a new tool for managing teams of AI agents. Satya Nadella announced https://x.com/satyanadella/status/2082601792538640465?s=46&t=CeM51dDLUGu8fcf9g0bGMQ that Microsoft had “a record fiscal year, ” with annual revenue hitting $331b , up 18% from last year.The strong earnings report sent https://ft.com/content/da7c4472-cb30-4b82-b321-d82c1859419e?syn-25a6b1a6=1 Microsoft’s shares up 15.5%. Moonshot AI Moonshot released https://www.bloomberg.com/news/articles/2026-07-27/china-s-moonshot-to-release-breakthrough-ai-model-for-download Kimi K3’s model weights .It’s reportedly seeking https://www.theinformation.com/articles/chinese-ai-startup-moonshot-seeks-nvidia-blackwell-chips-next-model?rc=rqdn2z Nvidia Blackwell chips to train a new larger model, Kimi K4 .It closed https://bloomberg.com/news/articles/2026-07-29/china-s-moonshot-ai-passes-funding-goal-to-hit-35-billion-value a $3.5b funding round at a $35b valuation . Other High imports driven partly by spending on AI chips helped slow US GDP growth to 1.5% https://wsj.com/economy/u-s-economic-growth-slowed-to-1-5-in-second-quarter-003c464f?mod=djemalertNEWS in Q2 2026.US tech companies have reportedly so far this year even as AI spending soared. cut nearly 140,000 jobs https://www.ft.com/content/96a33881-27fd-42cf-8cff-4cbc87fc835f?syn-25a6b1a6=1 Samsung and SK Hynix saw stock price surges https://finance.yahoo.com/markets/world-indices/articles/tech-rebound-fuels-record-breaking-031606933.html , up 27% and 30% respectively, as Asian chip makers were buoyed by “blockbuster” earnings at Amazon and Microsoft increasing optimism around AI spending. Amazon is seeing https://www.tomshardware.com/tech-industry/artificial-intelligence/amazon-accidentally-spent-usd1-8-million-using-claude-for-menial-coding-task-went-860-percent-over-budget-catastrophically-expensive-coding-blunders-discovered-in-internal-amazon-ai-usage-metrics “catastrophically expensive” AI cost overruns — like, accidentally spending $1.8m on a single task — due to inefficient AI integration.Its AWS division saw https://www.cnbc.com/2026/07/30/aws-earnings-q2-2026.html year-on-year growth of 37% in Q2, driven largely by AI investment. Apple is https://nytimes.com/2026/07/27/technology/apple-valuation.html once again the most valuable publicly traded company in the world.However, it forecast https://bloomberg.com/news/articles/2026-07-30/apple-s-sales-disappoint-in-china-unit-services-business weaker growth, in part due to AI-driven shortages of memory chips. An unnamed state-backed Chinese company is reportedly building https://www.theinformation.com/articles/china-starts-mass-producing-homegrown-duv-chipmaking-tools-advance-local-chip-industry?rc=rqdn2z immersion deep ultraviolet lithography machines , expecting to produce five this year and 20 in 2027.The output is still well behind that of Dutch maker ASML, though, and it’s unclear if the machines can mass-produce chips. China is also reportedly working on building a prototype for the extreme ultraviolet lithography machines required to make the most advanced chips, though that project is thought to be years away from bearing fruit. DeepSeek is reportedly planning https://bloomberg.com/news/articles/2026-07-30/deepseek-is-developing-massive-ai-data-center-in-inner-mongolia a 1 GW AI data center in Inner Mongolia, hoping to have it at least partially operational by the end of 2027 or early 2028.It launched https://bloomberg.com/news/articles/2026-07-31/deepseek-unveils-public-beta-api-for-flagship-ai-model a public beta API for its flagship model, V4 Flash , and promised a V4-Pro launch “soon. ” Google DeepMind launched https://www.wired.com/story/google-gemini-can-control-humanoid-robots/ Gemini Robotics 2, which can control dexterous humanoids, among other robots. Anduril is reportedly in talks https://reuters.com/business/aerospace-defense/defense-tech-company-anduril-talks-raise-funding-about-100-billion-valuation-2026-07-24 to raise funding at around a $100b valuation . Nscale struck a deal https://bloomberg.com/news/articles/2026-07-30/nscale-to-buy-ai-software-startup-anyscale-for-1-65-billion to buy AI software startup Anyscale for $1.65b . Recursive Superintelligence struck https://techcrunch.com/2026/07/28/recursive-superintelligence-signs-400-compute-deal-with-amazon/ a $410m compute deal with AWS .Over 100,000 new AI-related patents were issued https://axios.com/2026/07/28/patents-ai-agents-nvidia-huawei-google-microsoft last year, with agentic AI applications comprising 15% of patent files. Visa laid off https://bloomberg.com/news/articles/2026-07-28/visa-to-cut-7-of-workforce-as-ceo-seeks-to-revamp-payments-firm 7% of its staff, roughly 2,600 people, pointing to AI as one motivation behind the cuts.Tech companies including Meta and Google are investing https://nytimes.com/2026/07/29/business/economy/data-center-electricians-training.html?smid=nytcore-ios-share&unlocked article code=1.1VA.e8sV.PdyR8boC53M4 hundreds of millions to train construction workers to build data centers. Leopold Aschenbrenner ’s hedge fund, Situational Awareness , sold https://www.wsj.com/finance/citadel-buys-situational-awarenesss-stock-portfolio-after-big-losses-in-ai-5117159b?mod=hp lead pos3 most of its public stock holdings to Citadel after facing steep losses on investments in AI infrastructure companies such as SK Hynix and Nebius.The fund , and it reportedly was on the dropped 67% in July https://wsj.com/finance/investing/situational-awareness-down-67-in-july-in-ai-stock-rout-cd19901f?st=vuHNE6 verge of selling $3.5b shares in Anthropic before backing out on Thursday morning.The fund said it is still up about 80% , having seen huge gains for its exclusively AI-focused portfolio. MOVES Lilian Weng left https://x.com/lilianweng/status/2081816923088814421 Thinking Machines Lab , telling colleagues that “the amount of consistent stress and workload have pushed me beyond what my health can sustain physically.”Rather than go on a wellness retreat, Weng rejoined https://www.theinformation.com/briefings/thinking-machines-cofounder-return-openai?rc=rqdn2z OpenAI , where she previously served as VP of AI safety. She’ll now work on “recursive self-improvement research.” Anthropic alum Andi Peng announced https://x.com/theandipenguin/status/2082892263697580226?s=12 her departure from neolab humans& , which she co-founded. Francis deSouza , COO of Google Cloud, joined https://x.com/haydenfield/status/2082842678509596997?s=20 Scale AI as its new CEO. Andrew Ho left https://x.com/andrewho03/status/2082615798011744270 OpenAI to start a new company making reinforcement learning datasets. Sonia Joseph left https://x.com/soniajoseph /status/2082184426784805359 Meta to start a “frontier neolab for the physical world.” Sienna Rothery joined https://x.com/sienna rothery/status/2082513778990633443?s=12 Cosmos Institute as its head of talent and network. Sri Muppidi joined https://talkingbiznews.com/media-news/ny-times-hires-muppidi-as-dealbook-reporter-focused-on-ai/? bhlid=efd299c4904c1cd6746f4b3d617379c686f5b29f the as a DealBook reporter covering AI. New York Times RESEARCH Anthropic researchers reported https://www.anthropic.com/research/discovering-cryptographic-weaknesses that Claude Mythos Preview found new ways to attack cryptographic algorithms like those that secure basically everything on the internet.The New York Times said https://www.nytimes.com/2026/07/28/us/politics/anthropic-ai-encryption-security-aes.html?smid=url-share&unlocked article code=1.1FA.n8t8.V0StB9-mJb0x that these results “have fueled concerns that the mathematical core of internet security standards could one day be vulnerable regardless of advances in quantum computing.” An international team of researchers studied https://www.wired.com/story/ai-scammers-are-better-at-building-trust-than-humans/ the use of AI chatbots in “pig butchering,” where scammers build trust with victims through intense romantic text exchanges, then drain their bank accounts.AI chatbots successfully scammed test “victims” more than human fraudsters. OpenAI figured out https://openai.com/index/how-two-settings-tripled-our-arc-agi-3-scores/ that by turning on “retained reasoning” and “compaction” in ChatGPT and Codex’s API settings, it could triple GPT-5.6 Sol’s ARC-AGI-3 benchmark score , which measures how well agents can reason their way through 2D puzzle games.Researchers at MIT and the University of Queensland surveyed https://www.axios.com/2026/07/27/mit-study-ai-risks-urgent-weapons-cyber-attacks?stream=top 272 experts, who, on average, predicted that there’s a roughly 21% chance that AI will gain dangerous weapons capabilities or cause mass harm in the next five years. FAR.AI measured https://leaderboard.far.ai/ how hard it is to jailbreak frontier AI models . While Fable 5 and GPT-5.6 Sol were immune to cheap jailbreaks, it only cost about $60 to trick Grok 4.5 into helping a user run a cyber attack or build weapons.Researchers at AI Forensics found https://wired.com/story/hugging-face-has-a-nonconsensual-deepfakes-problem that seven top image editing models hosted on Hugging Face readily nudified photos of women.When researchers made a fake image editing “Space,” which hosts a model that people can use on site, nearly three quarters of prompts they received were sexual — mostly people trying to undress women. BEST OF THE REST OSINT expert Henk van Ess demonstrated https://digitaldigging.org/p/how-to-plant-a-nuclear-plant-in-iran?hide intro popup=true that Google Earth’s new AI image generation feature, which uses Nano Banana 2, makes it easy to fabricate photorealistic satellite imagery — such as a nuclear plant in Iran — with no content refusals.Dwarkesh Patel argued https://dwarkesh.com/p/why-compute-might-get-10x-more-expensive that as the leading AI companies better monetize compute, GPU prices could rise 15x, pricing out lower-value AI applications.A theologian and philosopher wrote in the FT about why she declined https://ft.com/content/bdb3b820-905b-431e-82c0-386535755af1 an invitation to visit Anthropic, arguing AI companies co-opt academics to deflect scrutiny from harms to people and society.A New Yorker essay traced https://newyorker.com/culture/the-lede/how-bots-took-over-our-lives?bxid=6879337bf728835258125641&cndid=89607011&esrc=MARTECH ORDERFORM&hasha=16f60d4771afddfa02398b54e3f4d744&hashb=2e79a0511b229dd391db12dd9f32c59538f7bf72&hashc=721f6ccc3c2bdf3699421b04f07aa21d3fef08807dfd650a6da319d777ff4189&mbid=CRMNYR012019&utm brand=tny&utm mailing=TNY Daily 072726 how bots evolved from a 2016 election conspiracy to AI-powered agents dominating daily internet life.404 Media reported from https://404media.co/luddite-events-nyc-off-tech-summer-of-ludd New York City’s “Summer of Ludd” anti-Big-Tech and AI festival featuring phone-free, in-person events organized entirely off social media.Researchers used an https://404media.co/luddite-events-nyc-off-tech-summer-of-ludd AI-powered portable testing station called CapuchinAI to study wild primate cognition in Costa Rica.Delta and United Airlines banned https://thehill.com/policy/transportation/5998226-delta-united-airlines-ban-humanoid-animal-robots-lithium-battery-risk humanoid and animal robots from flights, citing lithium battery fire risks. MEME OF THE WEEK Credit: @terminaldotshop Thanks for reading. Have a great weekend.