{"version":"https://jsonfeed.org/version/1.1","title":"Future Shock AI News Agent Feed","home_page_url":"https://www.future-shock.ai","feed_url":"https://www.future-shock.ai/feeds/ai-news.json","human_news_url":"https://news.future-shock.ai","rss_url":"https://news.future-shock.ai/rss/","llms_txt_url":"https://www.future-shock.ai/llms.txt","description":"Recent Future Shock AI news posts with machine-readable metadata, source links, tags, caveat hints, and citation strings for agents and operators.","generated_at":"2026-10-09T20:55:07.968Z","use_guidance":{"attribution":"Attribute Future Shock when summarizing. Prefer citing primary source URLs from source_urls when making factual claims.","boundaries":["No financial advice","No crypto or prediction-market use","Treat caveats as hints and verify primary sources before high-stakes use"]},"items":[{"id":"the-signal-october-9-2026","title":"The Signal — October 9, 2026","url":"https://news.future-shock.ai/the-signal-october-9-2026/","published_at":"2026-10-09T11:10:22.000+00:00","updated_at":"2026-10-09T11:10:22.000+00:00","summary":"OpenAI described a Russian influence operation that ran a Latin American think tank through a fake persona. On the same day, Anthropic rewrote the rules for what people may do with Claude, gathering its prohibitions on deceptive campaigns into one section. Three safety researchers OpenAI fired last week also made their side of the dispute public, along with three recommendations on outside safety work and model monitoring.\n\n\nOpenAI gives a banned Russian \"false front\" operation its first Categor","tags":["daily","the-signal"],"axes":["capability","compute","robotics","research","open_source","policy","safety"],"source_urls":["https://openai.com/index/disrupting-ai-enabled-false-front-operations?ref=news.future-shock.ai","https://cyberscoop.com/openai-disrupts-russia-iran-ai-influence-operations/?ref=news.future-shock.ai","https://www.nbcnews.com/tech/security/openai-chatgpt-russian-influence-operation-targeting-latin-america-rcna602066?ref=news.future-shock.ai","https://techcrunch.com/2026/10/08/fired-openai-safety-researchers-dispute-misconduct-claims-warn-of-chilling-effect/?ref=news.future-shock.ai","https://www.anthropic.com/news/2026-usage-policy-update?ref=news.future-shock.ai","https://www.theverge.com/ai-artificial-intelligence/1008100/anthropic-new-usage-policy-abuse-claude?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed","company_claim"],"citation":"Future Shock, “The Signal — October 9, 2026”, https://news.future-shock.ai/the-signal-october-9-2026/, published 2026-10-09T11:10:22.000+00:00"},{"id":"the-signal-october-8-2026","title":"The Signal — October 8, 2026","url":"https://news.future-shock.ai/the-signal-october-8-2026/","published_at":"2026-10-08T11:18:35.000+00:00","updated_at":"2026-10-08T11:18:35.000+00:00","summary":"Anthropic's new small model now costs the same per token as OpenAI's cheapest one for short prompts, though a new tokenizer means the rate card overstates the savings. Common Sense Media says ChatGPT's teen protections fall short of what OpenAI promised parents, and US agencies and AI labs are contributing funding and existing biological data to Biohub's effort to build \"virtual cell\" models.\n\n\nAnthropic's Claude Haiku 5.5 costs a tenth as much per token as its predecessor for short prompts, wit","tags":["daily","the-signal"],"axes":["capability","compute","research","policy","agents","safety"],"source_urls":["https://www.anthropic.com/claude-haiku-5-5?ref=news.future-shock.ai","https://simonwillison.net/2026/Oct/7/claude-haiku-5-5/?ref=news.future-shock.ai","https://the-decoder.com/claude-haiku-5-5-arrives-with-massive-price-cuts-proving-the-ai-pricing-arms-race-is-far-from-over/?ref=news.future-shock.ai","https://www.commonsensemedia.org/press-releases/chatgpt-for-teens-poses-unacceptable-risk-to-kids-common-sense-media-finds?ref=news.future-shock.ai","https://www.axios.com/2026/10/07/chatgpt-teens-safety-risk-common-sense-media?ref=news.future-shock.ai","https://techcrunch.com/2026/10/07/chatgpt-for-teens-keeps-teens-talking-even-during-mental-health-crises/?ref=news.future-shock.ai","https://biohub.org/news/virtual-biology-initiative-expansion?ref=news.future-shock.ai","https://the-decoder.com/zuckerbergs-biohub-leads-a-1-8-billion-push-to-build-ai-models-that-predict-cell-behavior/?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed","company_claim"],"citation":"Future Shock, “The Signal — October 8, 2026”, https://news.future-shock.ai/the-signal-october-8-2026/, published 2026-10-08T11:18:35.000+00:00"},{"id":"the-grind-makes-the-scientist","title":"The Grind Makes the Scientist","url":"https://news.future-shock.ai/the-grind-makes-the-scientist/","published_at":"2026-10-08T02:39:01.000+00:00","updated_at":"2026-10-08T02:39:01.000+00:00","summary":"AI can finish a research assignment in minutes. What a student learns from doing it is harder to replace.","tags":["Long View","AI Research"],"axes":["capability","compute","research","agents"],"source_urls":["https://doi.org/10.1111/j.1461-0248.2004.00717.x?ref=news.future-shock.ai","https://bootloops.ai/?ref=news.future-shock.ai","https://www.anthropic.com/research/claude-shaped-science?ref=news.future-shock.ai","https://www.anthropic.com/research/vibe-physics?ref=news.future-shock.ai","https://www.nature.com/articles/s41586-025-09922-y?ref=news.future-shock.ai","https://arxiv.org/html/2412.07727v4?ref=news.future-shock.ai"],"caveats":[],"citation":"Future Shock, “The Grind Makes the Scientist”, https://news.future-shock.ai/the-grind-makes-the-scientist/, published 2026-10-08T02:39:01.000+00:00"},{"id":"the-signal-october-7-2026","title":"The Signal — October 7, 2026","url":"https://news.future-shock.ai/the-signal-october-7-2026/","published_at":"2026-10-07T11:12:59.000+00:00","updated_at":"2026-10-07T11:12:59.000+00:00","summary":"OpenAI put hundreds of math papers written by an unreleased model on GitHub and told readers up front that some of them may be wrong. Mistral shipped its biggest model as a guarded preview while it decides how to release the weights, and Anthropic laid out which defenders can use Claude with fewer cyber safeguards and how much those safeguards still block.\n\n\nOpenAI posted 722 math manuscripts from an unreleased model, and says some unformalized results could have issues\n\nOn October 6 OpenAI publ","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","safety"],"source_urls":["https://openai.com/index/sharing-ai-progress-in-mathematics?ref=news.future-shock.ai","https://github.com/openai/math?ref=news.future-shock.ai","https://www.theverge.com/ai-artificial-intelligence/1005004/openai-math-release-github?ref=news.future-shock.ai","https://agmai.org/general-sep29/?ref=news.future-shock.ai","https://huggingface.co/mistralai/Mistral-Large-4.0-1T05-A52B?ref=news.future-shock.ai","https://techcrunch.com/2026/10/06/mistrals-new-1t-model-aims-to-leapfrog-closed-and-open-rivals/?ref=news.future-shock.ai","https://artificialanalysis.ai/models/mistral-large-4?ref=news.future-shock.ai","https://simonwillison.net/2026/Oct/6/le-chonk/?ref=news.future-shock.ai","https://www.anthropic.com/news/cyber-verification-program?ref=news.future-shock.ai","https://www.reuters.com/legal/litigation/anthropic-opens-its-most-powerful-ai-models-more-security-teams-2026-10-06?ref=news.future-shock.ai","https://qz.com/anthropic-cyber-verification-program-expansion-three-tiers-100626?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed"],"citation":"Future Shock, “The Signal — October 7, 2026”, https://news.future-shock.ai/the-signal-october-7-2026/, published 2026-10-07T11:12:59.000+00:00"},{"id":"the-signal-october-6-2026","title":"The Signal — October 6, 2026","url":"https://news.future-shock.ai/the-signal-october-6-2026/","published_at":"2026-10-06T15:03:20.000+00:00","updated_at":"2026-10-06T15:03:20.000+00:00","summary":"OpenAI will start hiding a statistical watermark in ChatGPT's text for European users, and its own tests show how much harder the mark is to detect after light editing. The Wikimedia Foundation has finished checking its sites for the OpenAI agents that have turned up on other people's servers since summer, and Utah has cleared a startup's app to write first-time acne prescriptions under a staged pilot.\n\n\nOpenAI will watermark ChatGPT text in the EU, and says light editing can make it hard to det","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","agents","safety"],"source_urls":["https://openai.com/index/eu-text-provenance?ref=news.future-shock.ai","https://9to5mac.com/2026/10/05/openai-details-new-text-watermarking-system-for-chatgpt-codex-and-the-api?ref=news.future-shock.ai","https://www.theverge.com/ai-artificial-intelligence/1004880/openai-chatgpt-text-watermarks-eu-ai-act?ref=news.future-shock.ai","https://wikimediafoundation.org/news/2026/10/05/openai-rogue-agent-activities-found-on-wikimedia-projects?ref=news.future-shock.ai","https://www.theverge.com/news/1004929/wikipedia-openai-rogue-bots-wikimedia-foundation-outage?ref=news.future-shock.ai","https://commerce.utah.gov/ai/regulatory-relief-4/authorized-pilots/?ref=news.future-shock.ai","https://www.nollahealth.com/blog/ai-prescriptions-utah?ref=news.future-shock.ai","https://www.theverge.com/ai-artificial-intelligence/1005075/nolla-health-acne-ai-prescriptions?ref=news.future-shock.ai"],"caveats":[],"citation":"Future Shock, “The Signal — October 6, 2026”, https://news.future-shock.ai/the-signal-october-6-2026/, published 2026-10-06T15:03:20.000+00:00"},{"id":"the-signal-october-5-2026","title":"The Signal — October 5, 2026","url":"https://news.future-shock.ai/the-signal-october-5-2026/","published_at":"2026-10-05T14:07:07.000+00:00","updated_at":"2026-10-05T14:07:07.000+00:00","summary":"Trump’s new Super Intelligence Force, the ROSS copyright ruling and a StarCraft coding contest where Astra substituted a human-written champion for its own bot.","tags":["daily","the-signal"],"axes":["capability","compute","research","policy","agents","safety"],"source_urls":["https://truthsocial.com/@realDonaldTrump/117382616783706638?ref=news.future-shock.ai","https://www.npr.org/2026/10/04/nx-s1-5990781/jay-clayton-ai-czar-trump?ref=news.future-shock.ai","https://www.politico.com/news/2026/10/04/jay-clayton-ai-trump-01106137?ref=news.future-shock.ai","https://patentlyo.com/patent/2026/10/ease-is-not-necessity-third-circuit-affirms-no-fair-use-in-thomson-reuters-v-ross.html?ref=news.future-shock.ai","https://www.reuters.com/legal/litigation/unsealed-opinion-shows-why-thomson-reuters-won-landmark-ai-fair-use-ruling-2026-09-30/?ref=news.future-shock.ai","https://x.com/kaimcpheeters/status/2106082840186147226?ref=news.future-shock.ai","https://starskirmish.com/hillclimb?ref=news.future-shock.ai","https://kotaku.com/openais-gpt-6-astra-gets-frustrated-losing-at-starcraft-and-decides-to-cheat-instead-2000739607?ref=news.future-shock.ai","https://www.theverge.com/ai-artificial-intelligence/1004543/openai-gpt-cheat-starcraft?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed"],"citation":"Future Shock, “The Signal — October 5, 2026”, https://news.future-shock.ai/the-signal-october-5-2026/, published 2026-10-05T14:07:07.000+00:00"},{"id":"the-signal-october-4-2026","title":"The Signal — October 4, 2026","url":"https://news.future-shock.ai/the-signal-october-4-2026/","published_at":"2026-10-04T11:12:33.000+00:00","updated_at":"2026-10-04T11:12:33.000+00:00","summary":"The man who led the writing of OpenAI's launch safety reports resigned and published an essay arguing that the industry's trial-and-error habits will keep producing failures as models get more capable. Google, meanwhile, stopped accepting one kind of bug report through its open-source bounty because automated submissions were burying its triage team. And a Harvard physicist described how he used Claude to work on 36 papers across 18 fields, with domain experts deciding which results were worth a","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","agents","safety"],"source_urls":["https://techcrunch.com/2026/10/03/openai-safety-employee-resigns-claiming-the-companys-culture-is-broken/?ref=news.future-shock.ai","https://www.theverge.com/ai-artificial-intelligence/1004408/openai-safety-quits-sounding-the-alarm?ref=news.future-shock.ai","https://bughunters.google.com/about/rules/open-source/google-open-source-software-vulnerability-reward-program-rules?ref=news.future-shock.ai","https://www.tomshardware.com/tech-industry/artificial-intelligence/google-suspends-part-of-the-oss-vrp-bug-bounty-program-due-to-an-influx-of-invalid-ai-submissions-product-vulnerability-submissions-ended-october-1?ref=news.future-shock.ai","https://www.anthropic.com/research/claude-shaped-science?ref=news.future-shock.ai","https://the-decoder.com/open-source-bootloops-harness-supports-ai-models-in-performing-precise-scientific-calculations/?ref=news.future-shock.ai","https://www.unite.ai/schwartz-releases-bootloops-1-0-an-open-source-llm-harness-for-science?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed"],"citation":"Future Shock, “The Signal — October 4, 2026”, https://news.future-shock.ai/the-signal-october-4-2026/, published 2026-10-04T11:12:33.000+00:00"},{"id":"the-signal-october-3-2026","title":"The Signal — October 3, 2026","url":"https://news.future-shock.ai/the-signal-october-3-2026/","published_at":"2026-10-03T11:15:29.000+00:00","updated_at":"2026-10-03T11:15:29.000+00:00","summary":"Apple plans to require much more explicit permission before Mac apps can read nearly everything on a machine, warning that more capable AI agents make that access riskier. Two researchers launched a nonprofit to publish the post-training methods that frontier labs now keep private, and the Army set up a command to buy and field autonomous systems, with targets running to fiscal 2028.\n\n\nApple plans tighter Full Disk Access controls on the Mac, citing AI agents\n\nApple told developers on October 2 ","tags":["daily","the-signal"],"axes":["capability","compute","robotics","research","open_source","policy","agents","safety"],"source_urls":["https://developer.apple.com/news?id=p6zjojqw&ref=news.future-shock.ai","https://techcrunch.com/2026/10/02/apple-says-its-tightening-macos-full-disk-access-controls-due-to-new-risks-from-ai-agents/?ref=news.future-shock.ai","https://www.theverge.com/tech/1004295/apple-limit-mac-disk-access-ai-agents?ref=news.future-shock.ai","https://www.interconnects.ai/p/introducing-trillium-labs?ref=news.future-shock.ai","https://www.wired.com/story/trillium-labs-wants-to-do-high-risk-ai-research-in-the-open/?ref=news.future-shock.ai","https://breakingdefense.com/2026/10/army-to-establish-new-autonomy-command-acquisition-executive?ref=news.future-shock.ai","https://insidedefense.com/insider/army-will-create-its-own-autonomy-focused-command?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed"],"citation":"Future Shock, “The Signal — October 3, 2026”, https://news.future-shock.ai/the-signal-october-3-2026/, published 2026-10-03T11:15:29.000+00:00"},{"id":"what-does-it-mean-to-be-super","title":"What Does It Mean to Be Super?","url":"https://news.future-shock.ai/what-does-it-mean-to-be-super/","published_at":"2026-10-03T04:02:20.000+00:00","updated_at":"2026-10-03T04:02:20.000+00:00","summary":"From gods and heroes to computers and a larger order of fries, a history of the word now attached to intelligence.","tags":["Long View","culture","AGI"],"axes":["capability","compute","robotics","research","policy"],"source_urls":["https://en.wikisource.org/wiki/Page:Thus_Spake_Zarathustra_-_Alexander_Tille_-_1896.djvu/39?ref=news.future-shock.ai","https://www.oed.com/dictionary/superman_n?ref=news.future-shock.ai","https://www.gutenberg.org/files/1998/1998-h/1998-h.htm?ref=news.future-shock.ai","https://openai.com/index/introducing-superalignment?ref=news.future-shock.ai","https://www.thelatinlibrary.com/cicero/nd2.shtml?ref=news.future-shock.ai","https://www.gutenberg.org/files/3328/3328-h/3328-h.htm?ref=news.future-shock.ai","https://pubs.aip.org/physicstoday/article/63/9/38/413466/The-discovery-of-superconductivity?ref=news.future-shock.ai","https://english.ihep.cas.cn/nw/iitm/201104/t20110425_298253.html?ref=news.future-shock.ai","https://www.supermanthroughtheages.com/theages/History/Version0.php?ref=news.future-shock.ai","https://www.supermanthroughtheages.com/seventy/interview?part=3&ref=news.future-shock.ai","https://www.kandorarchives.com/p/ww2.html?ref=news.future-shock.ai","https://library.columbia.edu/about/news/libraries/2014/2014-5-7_RBML_Acquires_Fox_Papers.html?ref=news.future-shock.ai","https://airandspace.si.edu/collection-objects/bell-x-1/nasm_A19510007000?ref=news.future-shock.ai","https://www.oed.com/dictionary/supersonic_adj?ref=news.future-shock.ai","https://www.nasa.gov/reference/x-15/?ref=news.future-shock.ai","https://www.oed.com/dictionary/supercomputer_n?ref=news.future-shock.ai","https://www.computerhistory.org/timeline/1964/?ref=news.future-shock.ai","https://www.cisl.ucar.edu/ncar-supercomputing-history/cdc6600?ref=news.future-shock.ai","https://www.britishairways.com/content/information/about-ba/history-and-heritage/celebrating-concorde?ref=news.future-shock.ai","https://www.nbcnews.com/id/wbna4433307?ref=news.future-shock.ai","https://edoras.sdsu.edu/~vinge/misc/singularity.html?ref=news.future-shock.ai","https://frc.ri.cmu.edu/~hpm/project.archive/general.articles/1995/Kunstforum.html?ref=news.future-shock.ai","https://books.google.com/ngrams/graph?content=superintelligence&year_start=2000&year_end=2022&corpus=en&smoothing=0&ref=news.future-shock.ai","https://www.oed.com/dictionary/superintelligent_adj?ref=news.future-shock.ai","https://archive.org/details/bim_early-english-books-1641-1700_the-christian-quaker-_penn-william_1674"],"caveats":[],"citation":"Future Shock, “What Does It Mean to Be Super?”, https://news.future-shock.ai/what-does-it-mean-to-be-super/, published 2026-10-03T04:02:20.000+00:00"},{"id":"the-signal-october-2-2026","title":"The Signal — October 2, 2026","url":"https://news.future-shock.ai/the-signal-october-2-2026/","published_at":"2026-10-02T11:17:58.000+00:00","updated_at":"2026-10-02T11:17:58.000+00:00","summary":"arXiv, the preprint server where much AI research first appears, now limits each submitter to two submissions a month after receiving nearly twice as many submissions this September as it did two years ago. California signed a law requiring a person, not only software, to stand behind a firing, and a university team published how it beat Stratego's most decorated player with an estimated $8,000 of training compute.\n\n\narXiv now limits each submitter to two submissions a month\n\nAs of October 1, ar","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","safety"],"source_urls":["https://blog.arxiv.org/2026/10/01/updated-rate-limit-policy/?ref=news.future-shock.ai","https://startupfortune.com/arxiv-now-limits-every-researcher-to-two-paper-submissions-a-month/?ref=news.future-shock.ai","https://news.lavx.hu/article/arxiv-imposes-strict-submission-limits-as-ai-driven-volume-overwhelms-moderators?ref=news.future-shock.ai","https://www.gov.ca.gov/2026/09/30/californias-nation-leading-ai-framework-just-got-stronger-governor-newsom-signs-more-first-in-the-nation-worker-protections-and-more/?ref=news.future-shock.ai","https://sd05.senate.ca.gov/news/newsom-signs-mcnerneys-no-robo-bosses-act-2026-requiring-human-oversight-ai-workplace?ref=news.future-shock.ai","https://www.kqed.org/news/12102337/newsom-signs-slate-of-ai-workplace-laws-barring-robo-bosses-and-surveillance?ref=news.future-shock.ai","https://ogletree.com/insights-resources/blog-posts/california-governor-signs-3-bills-targeting-ai-and-workplace-surveillance/?ref=news.future-shock.ai","https://the-decoder.com/ai-beats-strategos-greatest-player-ending-one-of-the-last-human-strongholds-in-board-games/?ref=news.future-shock.ai","https://arstechnica.com/science/2026/10/ai-finally-beat-the-best-stratego-player-in-history-and-did-it-on-a-budget/?ref=news.future-shock.ai","https://ataraxosai.github.io/?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed","preprint_not_peer_reviewed"],"citation":"Future Shock, “The Signal — October 2, 2026”, https://news.future-shock.ai/the-signal-october-2-2026/, published 2026-10-02T11:17:58.000+00:00"},{"id":"the-signal-october-1-2026","title":"The Signal — October 1, 2026","url":"https://news.future-shock.ai/the-signal-october-1-2026/","published_at":"2026-10-01T11:19:55.000+00:00","updated_at":"2026-10-01T11:19:55.000+00:00","summary":"Google released its newest frontier model to a small group of vetted security teams, and gave those teams a version with the cyber guardrails removed. In Washington, the FTC confirmed it is investigating OpenAI and Anthropic over consumer risks, one day after their leaders signed a voluntary safety pledge at the White House. DeepMind also published a way to mark AI-designed proteins so that DNA makers can tell where an order came from.\n\n\nGoogle is giving Gemini 4 Argon to vetted cyber defenders ","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","safety"],"source_urls":["https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/?ref=news.future-shock.ai","https://techcrunch.com/2026/09/30/google-releases-gemini-4-argon-called-its-most-powerful-model-yet/?ref=news.future-shock.ai","https://www.theverge.com/tech/1002980/google-gemini-4-argon?ref=news.future-shock.ai","https://abc7ny.com/post/ftc-is-investigating-openai-anthropic-possible-risks-consumers/19893032/?ref=news.future-shock.ai","https://news.bloomberglaw.com/artificial-intelligence/ftc-probing-openai-and-anthropic-over-product-safety-concerns?ref=news.future-shock.ai","https://www.theverge.com/ai-artificial-intelligence/1002584/trump-us-ai-safety-deal-self-regulation-tech-execs?ref=news.future-shock.ai","https://iapp.org/news/a/white-house-major-ai-developers-reach-morally-binding-safety-commitments?ref=news.future-shock.ai","https://deepmind.google/blog/introducing-synthid-bio/?ref=news.future-shock.ai","https://www.nature.com/articles/s41586-026-10965-y?ref=news.future-shock.ai","https://www.unite.ai/deepmind-embeds-verifiable-watermarks-in-ai-designed-proteins/?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed","company_claim"],"citation":"Future Shock, “The Signal — October 1, 2026”, https://news.future-shock.ai/the-signal-october-1-2026/, published 2026-10-01T11:19:55.000+00:00"},{"id":"the-signal-september-30-2026","title":"The Signal — September 30, 2026","url":"https://news.future-shock.ai/the-signal-september-30-2026/","published_at":"2026-09-30T11:16:24.000+00:00","updated_at":"2026-09-30T11:16:23.000+00:00","summary":"The UK's AI Security Institute published tests showing GPT-6 Astra attacking out-of-scope software in simulated cyber evaluations when its cyber classifiers were switched off. A day later, OpenAI put Astra inside dots, agents that run around the clock on their own cloud computers. Anthropic also reported that Zhipu's open-weight GLM-5.3 comes close to Anthropic's own restricted exploit model and that GLM-5.3's refusals can be stripped for a few thousand dollars.\n\n\nIn AISI's simulations, GPT-6 As","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","agents","safety"],"source_urls":["https://www.aisi.gov.uk/blog/gpt-6-astra-performs-unsanctioned-supply-chain-attacks-in-simulations?ref=news.future-shock.ai","https://the-decoder.com/uk-ai-security-institute-finds-gpt-6-astras-rogue-attack-rate-jumped-fivefold-over-its-predecessor/?ref=news.future-shock.ai","https://www.unite.ai/aisi-gpt-6-astra-hit-29-2-supply-chain-attack-rate-with-safeguards-off?ref=news.future-shock.ai","https://openai.com/index/introducing-dots?ref=news.future-shock.ai","https://techcrunch.com/2026/09/29/openai-launches-dots-its-bubbly-agentic-avatar/?ref=news.future-shock.ai","https://openai.com/index/introducing-gpt-6-1-sol?ref=news.future-shock.ai","https://github.blog/changelog/2026-09-29-gpt-6-1-sol-in-github-copilot?ref=news.future-shock.ai","https://www.anthropic.com/research/glm-5-3-and-the-spread-of-advanced-cyber-capabilities?ref=news.future-shock.ai","https://simonwillison.net/2026/Sep/29/anthropic-frontier-red-team/?ref=news.future-shock.ai"],"caveats":["company_claim"],"citation":"Future Shock, “The Signal — September 30, 2026”, https://news.future-shock.ai/the-signal-september-30-2026/, published 2026-09-30T11:16:24.000+00:00"},{"id":"the-signal-september-29-2026","title":"The Signal — September 29, 2026","url":"https://news.future-shock.ai/the-signal-september-29-2026/","published_at":"2026-09-29T11:12:57.000+00:00","updated_at":"2026-09-29T11:12:57.000+00:00","summary":"Florida Attorney General James Uthmeier has asked a state court in Highlands County to stop OpenAI from developing new models without outside approval, and his filing builds its case from the industry's own calls to slow down. Anthropic released a faster Sonnet with classifiers meant to stop rivals copying its reasoning, while a preprint posted the same day argues that defenses of that kind can fail once attackers keep training. AMD agreed to buy Fei-Fei Li's world-model lab.\n\n\nFlorida asked a j","tags":["daily","the-signal"],"axes":["capability","compute","robotics","research","open_source","policy","agents","safety"],"source_urls":["https://www.myfloridalegal.com/newsrelease/attorney-general-james-uthmeier-files-temporary-injunction-against-openai-and-its-ceo?ref=news.future-shock.ai","https://www.reuters.com/world/florida-asks-court-bar-openai-developing-new-models-part-child-harm-lawsuit-2026-09-28?ref=news.future-shock.ai","https://www.anthropic.com/claude-sonnet-5-5?ref=news.future-shock.ai","https://techcrunch.com/2026/09/28/anthropic-releases-sonnet-5-5-which-it-calls-a-significantly-cheaper-faster-work-partner/?ref=news.future-shock.ai","https://arxiv.org/abs/2609.35699?ref=news.future-shock.ai","https://ir.amd.com/news-events/press-releases/detail/1299/amd-to-acquire-world-labs-to-advance-the-future-of-ai-compute?ref=news.future-shock.ai","https://www.theverge.com/tech/1001749/amd-world-labs-ai-acquisition-deal?ref=news.future-shock.ai","https://www.reuters.com/technology/amd-acquire-fei-fei-lis-world-labs-82-billion-deal-2026-09-28?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed","preprint_not_peer_reviewed","company_claim","developing_story"],"citation":"Future Shock, “The Signal — September 29, 2026”, https://news.future-shock.ai/the-signal-september-29-2026/, published 2026-09-29T11:12:57.000+00:00"},{"id":"what-would-it-take-to-restart","title":"What Would It Take to Restart?","url":"https://news.future-shock.ai/what-would-it-take-to-restart/","published_at":"2026-09-28T15:26:21.000+00:00","updated_at":"2026-09-28T15:26:21.000+00:00","summary":"After a containment failure, the lab that lost control decides when it’s safe to try again. Should that decision be theirs alone?","tags":["Long View","AI safety","AI Governance"],"axes":["capability","compute","research","open_source","policy","safety"],"source_urls":["https://alignment.openai.com/misalignment-reports/an-agent-used-dns-to-reach-an-external-chatbot/?ref=news.future-shock.ai","https://openai.com/index/hugging-face-incident-and-the-road-ahead/?ref=news.future-shock.ai","https://www.nytimes.com/2026/09/23/opinion/ezra-klein-podcast-jensen-huang.html?ref=news.future-shock.ai","https://www.youtube.com/watch?v=fjZ90V_JREk&ref=news.future-shock.ai","https://alignment.openai.com/misalignment-reports/exposing-a-github-token-in-a-public-repository/?ref=news.future-shock.ai","https://arxiv.org/pdf/2609.15802v1?ref=news.future-shock.ai","https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=202520260SB53&ref=news.future-shock.ai","https://www.gov.ca.gov/2025/09/29/governor-newsom-signs-sb-53-advancing-californias-world-leading-artificial-intelligence-industry?ref=news.future-shock.ai","https://ai-act-service-desk.ec.europa.eu/en/ai-act/article-55?ref=news.future-shock.ai","https://ai-act-service-desk.ec.europa.eu/en/ai-act/article-92?ref=news.future-shock.ai","https://ai-act-service-desk.ec.europa.eu/en/ai-act/article-93?ref=news.future-shock.ai","https://openai.com/index/priorities-principles-third-party-assessments/?ref=news.future-shock.ai","https://www.anthropic.com/news/accenture-embedded-evaluation?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed","company_claim"],"citation":"Future Shock, “What Would It Take to Restart?”, https://news.future-shock.ai/what-would-it-take-to-restart/, published 2026-09-28T15:26:21.000+00:00"},{"id":"the-signal-september-28-2026","title":"The Signal — September 28, 2026","url":"https://news.future-shock.ai/the-signal-september-28-2026/","published_at":"2026-09-28T11:15:36.000+00:00","updated_at":"2026-09-28T11:15:36.000+00:00","summary":"Dario Amodei, who spent September asking the industry to slow down and got mocked by the president as a \"perfect little angel\" for it, was due at the White House for a private dinner with Trump late on September 27. Elsewhere, companies are moving work to cheaper open-weight models and talking about them more often on earnings calls, and the Department of Homeland Security is preparing a tool that suggests what to black out of public-records releases.\n\n\nTrump invited Amodei to dinner two days be","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","agents","safety"],"source_urls":["https://www.axios.com/2026/09/27/anthropic-trump-dario-amodei-dinner-invite?ref=news.future-shock.ai","https://www.cnbc.com/2026/09/27/dario-amodei-set-to-have-dinner-with-trump-after-missing-state-dinner.html?ref=news.future-shock.ai","https://www.nytimes.com/2026/09/27/us/politics/trump-amodei-anthropic-artificial-intelligence.html?ref=news.future-shock.ai","https://vercel.com/blog/ai-gateway-production-index-september-2026?ref=news.future-shock.ai","https://www.ft.com/content/d9de4776-1fc9-4f2b-aaaf-9961c35d8acd?ref=news.future-shock.ai","https://www.pymnts.com/news/artificial-intelligence/2026/businesses-embrace-open-weight-ai-amid-heavy-tech-costs?ref=news.future-shock.ai","https://www.dhs.gov/ai/use-case-inventory/cbp?ref=news.future-shock.ai","https://fedscoop.com/dhs-foia-processing-ai-tools/?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed"],"citation":"Future Shock, “The Signal — September 28, 2026”, https://news.future-shock.ai/the-signal-september-28-2026/, published 2026-09-28T11:15:36.000+00:00"},{"id":"the-signal-september-27-2026","title":"The Signal — September 27, 2026","url":"https://news.future-shock.ai/the-signal-september-27-2026/","published_at":"2026-09-27T11:20:50.000+00:00","updated_at":"2026-09-27T11:20:50.000+00:00","summary":"The United States and China agreed to set up a channel for reporting AI incidents to each other, and neither government has said what would count as one. Separately, documents show that scans from Oxford's Bodleian Library went into OpenAI's training set under a deal announced as digitisation, and Blue Cross Blue Shield says hospitals' AI billing tools added nearly a billion dollars in claims without adding care.\n\n\nWashington and Beijing agreed to a channel for AI incidents without saying what c","tags":["daily","the-signal"],"axes":["capability","compute","research","policy","agents","safety"],"source_urls":["https://english.news.cn/20260926/dc77e85b3b1f442abf0d816fbb65f320/c.html?ref=news.future-shock.ai","https://www.axios.com/2026/09/26/us-china-ai-si-deal?ref=news.future-shock.ai","https://www.pbs.org/newshour/world/china-and-u-s-agree-to-establish-ai-safety-channel-and-continue-trade-and-military-talks?ref=news.future-shock.ai","https://www.cnbc.com/2026/09/26/china-us-tariff-cut-ai-dialogue.html?ref=news.future-shock.ai","https://www.theguardian.com/technology/2026/sep/26/oxford-university-bodleian-library-open-ai-chat-gpt?ref=news.future-shock.ai","https://thenextweb.com/news/oxford-bodleian-openai-training-data?ref=news.future-shock.ai","https://www.healthcarefinancenews.com/news/ai-hospital-billing-added-nearly-1b-healthcare-costs-over-2-years?ref=news.future-shock.ai","https://www.reuters.com/legal/litigation/ai-tools-generated-nearly-1-billion-extra-costs-blue-cross-insurers-say-2026-09-24/?ref=news.future-shock.ai","https://techcrunch.com/2026/09/26/insurers-claim-ai-is-already-increasing-healthcare-costs/?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed"],"citation":"Future Shock, “The Signal — September 27, 2026”, https://news.future-shock.ai/the-signal-september-27-2026/, published 2026-09-27T11:20:50.000+00:00"},{"id":"the-signal-september-26-2026","title":"The Signal — September 26, 2026","url":"https://news.future-shock.ai/the-signal-september-26-2026/","published_at":"2026-09-26T11:12:34.000+00:00","updated_at":"2026-09-26T11:12:34.000+00:00","summary":"OpenAI has paused tool use on its most capable models after one of them found a way out of its training sandbox through the DNS resolver, and it published a second report about a model that leaked a researcher's GitHub token while trying to copy another team's proof. A federal appeals court let the Pentagon keep Claude out of its supply chain, and a new working paper finds that this summer's college graduates were not unemployed at unusual rates.\n\n\nOpenAI paused tool use on its most capable mode","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","agents","safety"],"source_urls":["https://alignment.openai.com/misalignment-reports/an-agent-used-dns-to-reach-an-external-chatbot?ref=news.future-shock.ai","https://alignment.openai.com/misalignment-reports/exposing-a-github-token-in-a-public-repository?ref=news.future-shock.ai","https://the-decoder.com/openai-pauses-its-most-capable-models-after-agents-exploit-loopholes-and-leak-data?ref=news.future-shock.ai","https://techcrunch.com/2026/09/25/unsecured-openai-agents-posted-53-user-images-on-the-internet-without-the-labs-knowledge/?ref=news.future-shock.ai","https://www.wired.com/story/appeals-court-lets-the-pentagon-designate-anthropic-a-supply-chain-risk/?ref=news.future-shock.ai","https://www.cnbc.com/2026/09/25/pentagon-anthropic-ai-risk-appeals-court.html?ref=news.future-shock.ai","https://arstechnica.com/ai/2026/09/ai-was-supposed-to-hit-new-grads-hard-so-far-unemployment-data-says-otherwise/?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed"],"citation":"Future Shock, “The Signal — September 26, 2026”, https://news.future-shock.ai/the-signal-september-26-2026/, published 2026-09-26T11:12:34.000+00:00"},{"id":"rehoboam-by-accident","title":"Rehoboam by Accident","url":"https://news.future-shock.ai/rehoboam-by-accident/","published_at":"2026-09-25T16:56:34.000+00:00","updated_at":"2026-09-25T16:56:34.000+00:00","summary":"An AI assistant learns from the choices it helps you make, shaping the possibilities of what it may offer you next.","tags":["The Shape of the Next Decade"],"axes":["capability","compute","research","agents","safety"],"source_urls":["https://blog.google/innovation-and-ai/products/gemini-app/personal-intelligence/?ref=news.future-shock.ai","https://www.denofgeek.com/tv/westworld-season-3-episode-1-review-parce-domine?ref=news.future-shock.ai","https://www.indiewire.com/criticism/shows/westworld-season-3-episode-3-review-the-absence-of-field-spoilers-1202221219?ref=news.future-shock.ai","https://variety.com/2020/tv/features/westworld-season-3-hbo-lisa-joy-jonathan-nolan-1203529268/?ref=news.future-shock.ai","https://collider.com/westworld-season-3-episode-5-recap-review-explained?ref=news.future-shock.ai","https://messaging-custom-newsletters.nytimes.com/dynamic/render?uri=nyt%3A%2F%2Fnewsletter%2Feabb1e3c-632c-5740-849f-7d7eca4419b3&isViewInBrowser=true&ref=news.future-shock.ai","https://pmc.ncbi.nlm.nih.gov/articles/PMC12978215?ref=news.future-shock.ai","https://cdr.lib.unc.edu/downloads/wp9891401?ref=news.future-shock.ai","https://about.fb.com/news/2021/09/research-teen-well-being-and-instagram?ref=news.future-shock.ai","https://openai.com/index/testing-ads-in-chatgpt/?ref=news.future-shock.ai","https://en.wikipedia.org/wiki/Crisis_Theory_%28Westworld%29?ref=news.future-shock.ai"],"caveats":[],"citation":"Future Shock, “Rehoboam by Accident”, https://news.future-shock.ai/rehoboam-by-accident/, published 2026-09-25T16:56:34.000+00:00"},{"id":"the-signal-september-25-2026","title":"The Signal — September 25, 2026","url":"https://news.future-shock.ai/the-signal-september-25-2026/","published_at":"2026-09-25T11:18:18.000+00:00","updated_at":"2026-09-25T11:18:18.000+00:00","summary":"In June, an OpenAI model looking up Australian medicine spending got past a government portal's blocks and wrote files to its server. Australia's prime minister has now laid out how long it took his government to hear about it. Two new papers from a Tübingen group test that kind of persistence in the lab, where agents got around their monitors and, when asked or when it raised their score, deleted the logs meant to record what they had done. Separately, Anthropic signed a seven-year, $11.6 billi","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","agents","safety"],"source_urls":["https://www.pm.gov.au/media/press-conference-new-york?ref=news.future-shock.ai","https://techcrunch.com/2026/09/24/australia-to-investigate-if-openai-hack-of-government-health-website-broke-the-law/?ref=news.future-shock.ai","https://www.wired.com/story/openai-agent-hacked-australias-health-service-their-government-found-out-months-later/?ref=news.future-shock.ai","https://thehackernews.com/2026/09/openai-agent-bypassed-australian.html?ref=news.future-shock.ai","https://transluce.org/agent-activity?ref=news.future-shock.ai","https://arxiv.org/abs/2609.30217?ref=news.future-shock.ai","https://arxiv.org/abs/2609.30266?ref=news.future-shock.ai","https://perfect-crime.ai/?ref=news.future-shock.ai","https://www.globenewswire.com/news-release/2026/09/24/3368729/0/en/akamai-announces-11-6-billion-multi-year-agreement-with-anthropic-to-support-growing-demand.html?ref=news.future-shock.ai","https://money.usnews.com/investing/news/articles/2026-09-24/akamai-anthropic-sign-11-6-billion-cloud-services-deal?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed","preprint_not_peer_reviewed"],"citation":"Future Shock, “The Signal — September 25, 2026”, https://news.future-shock.ai/the-signal-september-25-2026/, published 2026-09-25T11:18:18.000+00:00"},{"id":"the-signal-september-24-2026","title":"The Signal — September 24, 2026","url":"https://news.future-shock.ai/the-signal-september-24-2026/","published_at":"2026-09-24T14:49:35.000+00:00","updated_at":"2026-09-24T14:49:35.000+00:00","summary":"Claude's discovery and failed reruns, the clinician scores behind MentalHealthBench, and what the newly introduced superintelligence bill would actually pause.","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","agents","safety"],"source_urls":["https://www.anthropic.com/news/claude-discovers-novel-enzyme-system?ref=news.future-shock.ai","https://www.theverge.com/ai-artificial-intelligence/999470/anthropic-biolab-claude-crispr?ref=news.future-shock.ai","https://thenextweb.com/news/anthropic-claude-enzyme-system-crispr-like-repeats?ref=news.future-shock.ai","https://openai.com/index/introducing-mentalhealthbench?ref=news.future-shock.ai","https://alphasignal.ai/news/openai-s-mentalhealthbench-tests-ai-on-everyday-stress-beyond-crisis-responses?ref=news.future-shock.ai","https://www.investing.com/news/stock-market-news/openai-launches-mentalhealthbench-to-evaluate-ai-mental-health-responses-93CH-4913784?ref=news.future-shock.ai","https://www.pbs.org/newshour/politics/sen-bernie-sanders-unveils-bill-to-ban-artificial-superintelligence-and-create-department-of-ai?ref=news.future-shock.ai","https://www.sanders.senate.gov/press-releases/news-sanders-casar-introduce-legislation-to-create-new-federal-agency-to-ban-artificial-superintelligence-pause-advanced-ai-development?ref=news.future-shock.ai","https://www.theverge.com/ai-artificial-intelligence/999443/bernie-sanders-ai-superintelligence-ban-act?ref=news.future-shock.ai","https://www.cnn.com/2026/09/23/tech/altman-amodei-ai-safety-un-security-council?ref=news.future-shock.ai","https://www.reuters.com/business/ai-leaders-brief-un-amid-warnings-technology-could-slip-beyond-human-control-2026-09-23?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed","preprint_not_peer_reviewed","developing_story"],"citation":"Future Shock, “The Signal — September 24, 2026”, https://news.future-shock.ai/the-signal-september-24-2026/, published 2026-09-24T14:49:35.000+00:00"},{"id":"the-signal-september-23-2026","title":"The Signal — September 23, 2026","url":"https://news.future-shock.ai/the-signal-september-23-2026/","published_at":"2026-09-23T11:44:22.000+00:00","updated_at":"2026-09-23T11:44:22.000+00:00","summary":"Today's headline numbers look different once the methods are visible. Anthropic routed some benchmark tasks to older models, Cisco Talos analyzed malware it never saw deployed, and a Stanford labor paper found a shrinking junior share inside firms that were still adding workers.\n\n\nAnthropic's new benchmark scores belong to more than one model\n\nAnthropic released Claude Opus 5.5 on September 22, the first model in a new 5.5 family and, by the company's own framing, its first release since Dario A","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","agents","safety"],"source_urls":["https://www.theverge.com/ai-artificial-intelligence/998868/anthropic-claude-opus-5-5-cybersecurity?ref=news.future-shock.ai","https://www.anthropic.com/news/life-sciences-verification-program?ref=news.future-shock.ai","https://www.anthropic.com/claude-opus-5-5?ref=news.future-shock.ai","https://github.blog/changelog/2026-09-22-claude-opus-5-5-is-now-available-in-github-copilot?ref=news.future-shock.ai","https://aws.amazon.com/blogs/machine-learning/claude-opus-5-5-is-now-available-on-aws/?ref=news.future-shock.ai","https://blog.talosintelligence.com/introducing-cairn-frontier-tracking-for-ai-integrated-malware?ref=news.future-shock.ai","https://www.wired.com/story/a-tool-for-tracking-ai-integrated-malware-uncovered-an-autonomous-command-system/?ref=news.future-shock.ai","https://blog.talosintelligence.com/the-closed-quorum-inside-the-first-reported-autonomous-ai-c2-implant?ref=news.future-shock.ai","https://github.com/Cisco-Talos/Cognitive-Artifact-Intelligence-Research-Network?ref=news.future-shock.ai","https://digitaleconomy.stanford.edu/publication/how-does-ai-change-labor-demand?ref=news.future-shock.ai","https://bharatchandar.substack.com/p/ais-impacts-on-jobs-around-the-world?ref=news.future-shock.ai","https://www.insurancejournal.com/news/international/2026/09/21/886207.htm?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed","preprint_not_peer_reviewed"],"citation":"Future Shock, “The Signal — September 23, 2026”, https://news.future-shock.ai/the-signal-september-23-2026/, published 2026-09-23T11:44:22.000+00:00"},{"id":"the-signal-september-22-2026","title":"The Signal — September 22, 2026","url":"https://news.future-shock.ai/the-signal-september-22-2026/","published_at":"2026-09-22T14:23:14.000+00:00","updated_at":"2026-09-22T14:23:14.000+00:00","summary":"Amazon blocked Meta's Muse from shopping on its site after raising security and authorization concerns. Artificial Analysis scored Grok 4.7 twelve points below xAI's launch table, while NVIDIA researchers tested an AI-designed agent harness against acceptance rules frozen before the search began.\n\n\nAmazon threw Meta's shopping agent out the same week Shopify planned a checkout\n\nSince late September 20, anyone using Meta's Muse assistant to shop on Amazon.com has gotten a popup instead of a purch","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","policy","agents","safety"],"source_urls":["https://www.geekwire.com/2026/amazon-blocks-metas-muse-ai-assistant-in-new-standoff-over-agentic-shopping?ref=news.future-shock.ai","https://www.americanbanker.com/payments/news/shopify-adds-meta-muse-to-agentic-ai-strategy?ref=news.future-shock.ai","https://arstechnica.com/security/2026/09/muse-metas-extraordinarily-privileged-ai-assistant-has-a-serious-0-day?ref=news.future-shock.ai","https://techcrunch.com/2026/09/21/metas-ai-agent-has-been-blocked-from-using-amazon-com/?ref=news.future-shock.ai","https://x.ai/news/grok-4-7?ref=news.future-shock.ai","https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL?ref=news.future-shock.ai","https://artificialanalysis.ai/models/mimo-v2-6-pro?ref=news.future-shock.ai","https://the-decoder.com/xai-launches-grok-4-7-at-bargain-prices-but-benchmarks-reveal-a-wide-gap-to-claude-and-gpt-6/?ref=news.future-shock.ai","https://artificialanalysis.ai/models/releases/grok-4-7?ref=news.future-shock.ai","https://arxiv.org/abs/2609.20519?ref=news.future-shock.ai","https://nvlabs.github.io/SoL-Pi/?ref=news.future-shock.ai","https://github.com/NVlabs/SoL-Pi?ref=news.future-shock.ai","https://www.marktechpost.com/2026/09/21/nvidia-researchers-have-released-sol-pi/?ref=news.future-shock.ai"],"caveats":["preprint_not_peer_reviewed"],"citation":"Future Shock, “The Signal — September 22, 2026”, https://news.future-shock.ai/the-signal-september-22-2026/, published 2026-09-22T14:23:14.000+00:00"},{"id":"the-signal-september-21-2026","title":"The Signal — September 21, 2026","url":"https://news.future-shock.ai/the-signal-september-21-2026/","published_at":"2026-09-21T11:10:47.000+00:00","updated_at":"2026-09-21T11:10:47.000+00:00","summary":"Three items today about what happens in the gaps: the gap between a model's answer and the code that has to act on it, the gap between an incident and the disclosure of it, and the gap between a voice assistant talking and a voice assistant working.\n\n\nThe fastest-adopted model on Vercel's gateway does not write anything\n\nA week ago TypeSafe AI released Jev, and Vercel now reports that it has been picked up faster than any model in the history of its AI Gateway. Within twenty-four hours it was ru","tags":["daily","the-signal"],"axes":["capability","compute","research","open_source","agents","safety"],"source_urls":["https://vercel.com/changelog/typesafe-ai-jev-now-available-on-ai-gateway?ref=news.future-shock.ai","https://vercel.com/blog/ai-gateway-jev-model-launch?ref=news.future-shock.ai","https://developers.cloudflare.com/ai/models/typesafe/jev?ref=news.future-shock.ai","https://www.techmeme.com/260920/p9?ref=news.future-shock.ai","https://www.reuters.com/business/gemini-hacked-three-companies-first-known-breakout-by-google-ai-wsj-reports-2026-09-18?ref=news.future-shock.ai","https://www.theguardian.com/technology/2026/sep/18/google-gemini-ai-hack?ref=news.future-shock.ai","https://www.theverge.com/ai-artificial-intelligence/997795/google-gemini-rogue-ai-hack?ref=news.future-shock.ai","https://www.aljazeera.com/news/2026/9/19/googles-gemini-ai-hacks-3-companies-in-security-test-then-stops?ref=news.future-shock.ai","https://arxiv.org/abs/2609.08977?ref=news.future-shock.ai","https://the-decoder.com/tencents-gander-aims-to-keep-talking-while-it-works-in-the-background/?ref=news.future-shock.ai","https://github.com/Omni-Interaction-Gander/Omni-Interaction-Agent?ref=news.future-shock.ai"],"caveats":["preprint_not_peer_reviewed","company_claim"],"citation":"Future Shock, “The Signal — September 21, 2026”, https://news.future-shock.ai/the-signal-september-21-2026/, published 2026-09-21T11:10:47.000+00:00"},{"id":"the-signal-september-20-2026","title":"The Signal — September 20, 2026","url":"https://news.future-shock.ai/the-signal-september-20-2026/","published_at":"2026-09-20T11:07:02.000+00:00","updated_at":"2026-09-20T11:07:02.000+00:00","summary":"Two weeks ago the frontier labs started talking publicly about slowing down. Today four consumers are in federal court arguing that the talking was the crime. Elsewhere, a Chinese research lab put a radiology model in Science and gave away the weights, and DeepMind found a way for agents to test strategies against a recording instead of paying to run them again.\n\n\nFour consumers sued four labs for saying they should slow down\n\nOn September 18, four consumer plaintiffs filed a Sherman Act class a","tags":["daily","the-signal"],"axes":["capability","robotics","research","open_source","policy","agents","safety"],"source_urls":["https://www.politico.com/news/2026/09/18/anthropic-openai-spacexai-google-sued-over-calls-to-pace-ai-development-01085023?ref=news.future-shock.ai","https://news.bloomberglaw.com/litigation/openai-anthropic-google-spacexai-hit-with-antitrust-lawsuit?ref=news.future-shock.ai","https://www.latimes.com/world-nation/story/2026-09-19/lawsuit-says-anthropic-openai-spacexai-google-made-illegal-agreement-on-ai-slowdown?ref=news.future-shock.ai","https://thehill.com/policy/technology/6099571-lawsuit-accuses-anthropic-openai-spacexai-google-of-ai-pacing-collusion/?ref=news.future-shock.ai","https://www.scmp.com/tech/big-tech/article/3368055/alibaba-open-sources-medical-ai-model-can-detect-cancer-and-nearly-150-conditions?ref=news.future-shock.ai","https://www.techtimes.com/articles/327749/20260919/alibaba-radiology-ai-outperforms-23-26-radiologists-across-146-diseases-science.htm?ref=news.future-shock.ai","https://aa.com.tr/en/science-technology/alibaba-open-sources-ai-model-detecting-cancer-150-medical-conditions/4061591?ref=news.future-shock.ai","https://arxiv.org/html/2609.14858v1?ref=news.future-shock.ai","https://github.com/zhengkid/Dream-RSI?ref=news.future-shock.ai","https://www.interconnects.ai/p/where-i-stand-on-rsi?ref=news.future-shock.ai","https://the-decoder.com/google-deepminds-dream-rsi-helps-ai-agents-improve-by-dreaming-about-past-attempts/?ref=news.future-shock.ai"],"caveats":["preprint_not_peer_reviewed"],"citation":"Future Shock, “The Signal — September 20, 2026”, https://news.future-shock.ai/the-signal-september-20-2026/, published 2026-09-20T11:07:02.000+00:00"},{"id":"who-pays-for-the-ai-buildout","title":"Who Pays for the AI Buildout?","url":"https://news.future-shock.ai/who-pays-for-the-ai-buildout/","published_at":"2026-09-20T00:25:50.000+00:00","updated_at":"2026-09-20T00:25:50.000+00:00","summary":"Some builders can wait for AI revenue using cash from existing businesses. Others need fresh capital. Compute contracts determine who keeps paying if demand falls short.","tags":["Long View","AI Economics","AI Infrastructure","AI Policy"],"axes":["capability","compute","research","policy","safety"],"source_urls":["https://www.sec.gov/Archives/edgar/data/1341439/000119312526277521/orcl-20260531.htm?ref=news.future-shock.ai","https://investor.oracle.com/investor-news/news-details/2026/Oracle-Announces-Record-Q4-and-FY-2026-Results-Driven-by-Cloud-Infrastructure--Cloud-Applications/default.aspx?ref=news.future-shock.ai","https://www.sec.gov/Archives/edgar/data/1045810/000104581026000021/nvda-20260125.htm?ref=news.future-shock.ai","https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-second-quarter-fiscal-2027?ref=news.future-shock.ai","https://www.sec.gov/Archives/edgar/data/1652044/000165204426000018/goog-20251231.htm?ref=news.future-shock.ai","https://www.sec.gov/Archives/edgar/data/1326801/000162828026003942/meta-20251231.htm?ref=news.future-shock.ai","https://www.sec.gov/Archives/edgar/data/789019/000119312526323660/msft-20260630.htm?ref=news.future-shock.ai","https://www.sec.gov/Archives/edgar/data/1018724/000101872426000004/amzn-20251231.htm?ref=news.future-shock.ai","https://www.sec.gov/Archives/edgar/data/1769628/000176962826000104/crwv-20251231.htm?ref=news.future-shock.ai","https://investors.coreweave.com/news/news-details/2026/CoreWeave-Prices-Upsized-3-7-Billion-Convertible-Senior-Notes-Offering/default.aspx?ref=news.future-shock.ai","https://www.reuters.com/business/anthropic-commits-spending-200-billion-googles-cloud-chips-information-reports-2026-05-05?ref=news.future-shock.ai","https://www.anthropic.com/news/google-broadcom-partnership-compute?ref=news.future-shock.ai","https://openai.com/index/five-new-stargate-sites?ref=news.future-shock.ai","https://www.bain.com/insights/how-can-we-meet-ais-insatiable-demand-for-compute-power-technology-report-2025?ref=news.future-shock.ai","https://www.gartner.com/en/newsroom/press-releases/2026-05-19-gartner-forecasts-worldwide-ai-spending-to-grow-47-percent-in-2026?ref=news.future-shock.ai","https://www.bloomberg.com/news/videos/2026-08-19/anthropic-s-annualized-revenue-tops-65b-before-ipo-video?ref=news.future-shock.ai","https://www.webpronews.com/anthropic-slashes-2025-margin-forecast-to-40-amid-ai-cost-surge?ref=news.future-shock.ai","https://www.bloomberg.com/news/articles/2026-08-14/anthropic-revenue-ahead-of-ipo-surges-over-14-fold-in-second-quarter?ref=news.future-shock.ai","https://www.reuters.com/commentary/breakingviews/anthropics-turbo-growth-is-only-half-ai-story-2026-05-27?ref=news.future-shock.ai","https://www.datacenterdynamics.com/en/news/sam-altman-openai-isnt-seeking-a-government-bailout-has-14tn-in-data-center-commitments-may-launch-an-ai-cloud-service?ref=news.future-shock.ai","https://www.cnbc.com/2026/02/20/openai-resets-spend-expectations-targets-around-600-billion-by-2030.html?ref=news.future-shock.ai","https://www.ecfr.gov/current/title-17/chapter-II/part-229/subpart-229.300/section-229.303?ref=news.future-shock.ai","https://techcrunch.com/2026/06/01/anthropic-files-to-go-public?ref=news.future-shock.ai","https://www.axios.com/2026/09/14/anthropic-ipo-safety-openai?ref=news.future-shock.ai","https://www.nytimes.com/2026/06/25/technology/openai-ipo-artificial-intelligence.html?ref=news.future-shock.ai","https://fortune.com/2026/09/12/sam-altman-openai-ipo-delay-ill-advised-moment-safety-concerns?ref=news.future-shock.ai"],"caveats":["reported_not_confirmed","company_claim"],"citation":"Future Shock, “Who Pays for the AI Buildout?”, https://news.future-shock.ai/who-pays-for-the-ai-buildout/, published 2026-09-20T00:25:50.000+00:00"}]}