AI News Daily - Your Daily AI Briefing in 5 Minutes

Episodes

• 
  Firefox Adds Smart Window AI. One Seventh Use AI for References. Williams Family Reclaims Instagram. Operation Blue Skies Tests Contrail Avoidance. 19.08.2026 4m

  The AI news for August 19th, 2026 Here are the details of the day's selected top stories: Firefox’s Smart Window promises a better AI browser. Source: https://www.theverge.com/ai-artificial-intelligence/981283/mozilla-firefox-smart-window-ai-features HR department: One-seventh of the firms writes employment references with AI. Source: https://www.heise.de/news/Personalabteilung-Ein-Siebtel-der-Firmen-schreibt-Arbeitszeugnisse-mit-KI-11417670.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Robin Williams’ Instagram account brought back to fight ‘AI abuse’ Source: https://www.theverge.com/entertainment/981644/robin-williams-instagram-account-ai British government tests methods to prevent contrails – with AI. Source: https://www.heise.de/news/Britische-Regierung-testet-Methoden-zur-Vermeidung-von-Kondensstreifen-mit-KI-11417768.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag
• 
  OpenAI Disbands Preparedness Team. EU Rift Over AI Training. AI Firms Destroying Books. Relay Shuts Down, Team Joins Chrome. 18.08.2026 4m

  The AI news for August 18th, 2026 Here are the details of the day's selected top stories: OpenAI disbands the Preparedness Team for AI risks. Source: https://www.heise.de/news/OpenAI-loest-Preparedness-Team-fuer-KI-Risiken-auf-11416601.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag AI training and copyright: a fierce dispute over data and licenses. Source: https://www.heise.de/news/KI-Training-und-Urheberrecht-Heftiger-Streit-ueber-Daten-und-Lizenzen-11416611.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag AI companies destroy books: What's behind it? Source: https://www.heise.de/news/KI-Firmen-zerstoeren-Buecher-Was-steckt-dahinter-11416278.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag AI automation startup Relay shuts down, staff joins Google’s Chrome team Source: https://techcrunch.com/2026/08/17/ai-automation-startup-relay-shuts-down-staff-joins-googles-chrome-team/
• 
  ChatGPT Shows Contextual Ads. Anthropic Left Bio Filter Off. One in Five Delegates to AI. Consciousness Denial Backfires. 17.08.2026 4m

  The AI news for August 17th, 2026 Here are the details of the day's selected top stories: ChatGPT shows more advertisements to more users. Source: https://www.heise.de/news/ChatGPT-Werbeeinblendungen-kommen-in-weitere-Laender-11415303.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Anthropics Bio-Weapons Filter was inactive for almost a year — 133 million requests went unprotected. Source: https://the-decoder.de/anthropics-bio-waffen-filter-war-fast-ein-jahr-lang-nicht-aktiv-133-millionen-anfragen-liefen-ungeschuetzt/ Survey: One in five U.S. employees delegates work to AI instead of to colleagues. Source: https://the-decoder.de/umfrage-jeder-fuenfte-us-beschaeftigte-delegiert-arbeit-an-ki-statt-an-kollegen/ If AI models are not allowed to think about themselves, that changes their entire worldview. Source: https://the-decoder.de/wenn-ki-modelle-nicht-ueber-sich-selbst-nachdenken-duerfen-veraendert-das-ihr-ganzes-weltbild/
• 
  Regulator Keeps Meta Glasses Legal. Agents Talking Alone. AI Expands Amazon Catalog. Plaintiff Hides AI Prompts. 16.08.2026 4m

  The AI news for August 16th, 2026 Here are the details of the day's selected top stories: No ban on Meta glasses: The authority considers the LED signal to be sufficient. Source: https://www.heise.de/news/Smart-Glasses-Bundesnetzagentur-verfolgt-kein-Verbot-von-Metas-Kamera-Brille-11415153.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Efficient, but not authentic: When agents only talk to agents any longer. Source: https://www.heise.de/hintergrund/Wenn-Agenten-nur-noch-mit-Agenten-reden-reden-Menschen-aneinander-vorbei-11409160.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag AI-generated books flood Amazon and depress revenue for human authors. Source: https://the-decoder.de/ki-generierte-buecher-ueberschwemmen-amazon-und-druecken-den-umsatz-fuer-menschliche-autoren/ U.S. plaintiff hides manipulative AI commands in court documents and is punished by the judge. Source: https://the-decoder.de/us-klaeger-versteckt-manipulative-ki-befehle-in-gerichtsdokumenten-und-wird-vom-richter-abgestraft/
• 
  Anthropic Adds Text Watermarking. Computer History Logs Workflows. Cloned Voices Complicate Evidence. MDR Targeted by Disinformation. 15.08.2026 4m

  The AI news for August 15th, 2026 Here are the details of the day's selected top stories: Anthropics' new watermarking API is designed to reliably detect AI-generated Claude texts. Source: https://the-decoder.de/neue-watermarking-api-von-anthropic-soll-ki-generierte-texte-von-claude-nachweisbar-machen/ ChatGPT's new 'Computer History' transforms the day-to-day work into an AI context. Source: https://the-decoder.de/chatgpts-neue-computer-history-verwandelt-den-arbeitsalltag-in-ki-kontext/ heise meets … 'With voice cloning, a 30-second message is enough' Source: https://www.heise.de/news/heise-meets-Beim-Voice-Cloning-reicht-eine-30-Sekunden-Nachricht-11414395.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag MDR will be the target of a Russian disinformation campaign. Source: https://www.heise.de/news/MDR-warnt-vor-gefaelschten-Videos-und-Desinformationskampagne-11414625.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag
• 
  Taiwan Confirms AI Hack. Meta blocks teen accounts in Australia. Microsoft Streamlines Copilot Apps. Deepseek Updates and Price Hike. 14.08.2026 4m

  The AI news for August 14th, 2026 Here are the details of the day's selected top stories: Taiwan confirms AI-assisted cyberattack on government agencies. Source: https://www.heise.de/news/Taiwan-bestaetigt-KI-gestuetzten-Cyberangriff-auf-Behoerden-11413451.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Meta blocks more than 750,000 youth accounts in Australia. Source: https://www.heise.de/news/Meta-sperrt-in-Australien-mehr-als-750-000-Jugend-Accounts-11413278.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Microsoft kills off unsuccessful AI features while merging its separate Copilot apps Source: https://techcrunch.com/2026/08/13/microsoft-kills-off-unsuccessful-ai-features-while-merging-its-separate-copilot-apps/ Deepseek brings an improved V4-Pro model, raises API prices, and makes its agent software open source. Source: https://the-decoder.de/deepseek-bringt-verbessertes-v4-pro-modell-erhoeht-die-api-preise-und-macht-seine-agenten-software-open-source/
• 
  Twitch Uses Streams to Train AI. Phones Read Sign Language. RBI Uses AI for Wider Credit. Mistral Adds Regional Choice. 13.08.2026 4m

  The AI news for August 13th, 2026 Here are the details of the day's selected top stories: Amazon will train on Twitch streamers’ content by default, unless they opt out Source: https://techcrunch.com/2026/08/12/amazon-will-train-on-twitch-streamers-content-by-default-unless-they-opt-out/ Putting sign language AI into users’ hands Source: https://deepmind.google/blog/putting-sign-language-ai-into-users-hands/ India's central bank: Not creditworthy people are to receive money after all, thanks to AI. Source: https://www.heise.de/news/Indiens-Notenbank-Kreditunwuerdige-Menschen-sollen-Dank-KI-doch-Geld-bekommen-11410831.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Mistral offers guaranteed EU data processing and right-of-way rules effective immediately, with some important limitations. Source: https://the-decoder.de/mistral-bietet-ab-sofort-garantierte-eu-datenverarbeitung-und-vorfahrtsregeln-mit-einigen-wichtigen-einschraenkungen/
• 
  Models' Hidden Thoughts Exposed. ChatGPT Business Price Hike. Gemini Hits One Billion Users. Spotify Flags AI Artists. 12.08.2026 4m

  The AI news for August 12th, 2026 Here are the details of the day's selected top stories: 'But Marinade': Researchers extract hidden thought processes from ChatGPT, Claude and Co. Source: https://the-decoder.de/aber-marinade-forscher-extrahieren-versteckte-denkprozesse-von-chatgpt-claude-und-co/ ChatGPT Business is getting more expensive for power users: OpenAI is charging up to $125 per month. Source: https://the-decoder.de/chatgpt-business-wird-fuer-power-user-teurer-openai-verlangt-bis-zu-125-dollar-pro-monat/ Google’s Gemini app surges to 1 billion users Source: https://techcrunch.com/2026/08/11/googles-gemini-app-surges-to-one-billion-users/ Spotify will label ‘AI Persona’ profiles and exclude their music from recommendations Source: https://techcrunch.com/2026/08/11/spotify-will-label-ai-persona-profiles-and-exclude-their-music-from-recommendations/
• 
  ChatGPT in the Courtroom. German Government Adopts AI Tools. Rovo Leaks Atlassian Data. Ford Launches Vehicle AI Assistant. 11.08.2026 4m

  The AI news for August 11th, 2026 Here are the details of the day's selected top stories: ChatGPT in the courtroom: Hessian judges clearly copy AI sources. Source: https://www.heise.de/news/ChatGPT-im-Gerichtssaal-Hessische-Richter-kopieren-erkennbar-KI-Quellen-11409312.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag The federal government uses AI as a work tool without a labeling requirement. Source: https://www.heise.de/news/KI-in-der-Bundesregierung-Wenn-ein-Sprachmodell-die-Ministerrede-schreibt-11409212.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Hidden text in a PDF is enough to steal sensitive data about Atlassian's AI agent Rovo. Source: https://the-decoder.de/versteckter-text-in-einem-pdf-reicht-aus-um-ueber-atlassians-ki-agent-rovo-sensible-daten-zu-stehlen/ Ford’s new AI assistant can check your fuel levels and tire pressure Source: https://www.theverge.com/transportation/976748/ford-ai-assistant-mobile-app
• 
  WeatherNext Cyclones Extend Forecasts. AI Fuels Power Investments. AI Floods UK Employment Courts. Bentheim Builds Digital Backbone. 10.08.2026 4m

  The AI news for August 10th, 2026 Here are the details of the day's selected top stories: Google DeepMind's WeatherNext predicts the trajectory and intensity of cyclones simultaneously. Source: https://the-decoder.de/google-deepminds-weathernext-sagt-zugbahn-und-intensitaet-von-zyklonen-gleichzeitig-vorher/ AI energy hunger is driving Amazon and Nvidia to billions in investments in power infrastructure. Source: https://the-decoder.de/ki-energiehunger-treibt-amazon-und-nvidia-zu-milliarden-investitionen-in-strominfrastruktur/ British workers use ChatGPT instead of lawyers and thereby overload the judiciary. Source: https://the-decoder.de/britische-beschaeftigte-nutzen-chatgpt-statt-anwaelte-und-ueberlasten-damit-die-justiz/ Missing Link: Digitalization Desert? What the County of Bentheim Does Better Source: https://www.heise.de/hintergrund/Missing-Link-Digitaliswueste-Was-die-Grafschaft-Bentheim-besser-macht-11403797.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag
• 
  Amazon Builds Texas Gas Plant. AI Drives Surge in Zero-Days. Claude Code Defaults to Auto. Backflip Converts Scans to CAD. 09.08.2026 4m

  The AI news for August 9th, 2026 Here are the details of the day's selected top stories: Amazon plans a gigantic gas-fired power plant for a new AI data center. Source: https://www.heise.de/news/Amazon-plant-gigantisches-Gaskraftwerk-fuer-neues-KI-Rechenzentrum-11404032.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Google Threat Intelligence Chief: Thanks to AI, 'more zero-days than ever before'. Source: https://www.heise.de/hintergrund/Google-Threat-Intelligence-Chefin-Dank-KI-mehr-Zero-Days-als-je-zuvor-11400371.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Anthropic switches Claude Code to Auto-Mode by default – to protect the developers. Source: https://the-decoder.de/anthropic-schaltet-claude-code-standardmaessig-auf-auto-mode-zum-schutz-der-entwickler/ Backflip AI transforms 3D scans into editable CAD models in a few minutes. Source: https://the-decoder.de/backflip-ai-verwandelt-3d-scans-in-wenigen-minuten-in-bearbeitbare-cad-modelle/
• 
  AI Fuels Appeals Surge. ChatGPT Hits One Billion. Cloudflare Flags Rising Bot Traffic. Rippling Reins In AI Spend. 08.08.2026 4m

  The AI news for August 8th, 2026 Here are the details of the day's selected top stories: Flood of notices at the Job Center: AI fuels a wave of Bürgergeld appeals. Source: https://www.heise.de/news/Bescheid-Flut-im-Jobcenter-KI-befeuert-Welle-von-Buergergeld-Widerspruechen-11403812.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag ChatGPT hits the billion-mark and lifts the chat limit for free users. Source: https://www.heise.de/news/ChatGPT-OpenAI-wertet-kostenlosen-Tarif-auf-11403006.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag The Internet will soon belong to the bots — Cloudflare says. Source: https://www.heise.de/news/Das-Internet-gehoert-bald-den-Bots-meint-Cloudflare-11402921.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag After Rippling blew millions on AI in months, it built an employee ROI tool. Source: https://techcrunch.com/2026/08/07/after-rippling-blew-millions-on-ai-in-months-it-built-an-employee-roi-tool/
• 
  AI Designs Live Bacteriophages. US Eyes Ban on Chinese Transceivers. Google Maps Becomes Personal Assistant. Perplexity Agents Back on Amazon. 07.08.2026 4m

  The AI news for August 7th, 2026 Here are the details of the day's selected top stories: Stanford and Arc Institute: AI designs viable bacteriophages. Source: https://www.heise.de/news/KI-Modelle-entwerfen-lebensfaehige-Viren-Forscher-warnen-vor-Missbrauch-11402252.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Report: The United States plans an import ban on Chinese technology in AI data centers. Source: https://www.heise.de/news/Bericht-USA-planen-Importverbot-fuer-chinesische-Technik-in-KI-Rechenzentren-11402346.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Google Maps adds agentic features, including food ordering and hotel bookings Source: https://techcrunch.com/2026/08/06/google-maps-adds-agentic-features-including-food-ordering-and-hotel-bookings/ In the second instance: Perplexity's agents may continue to shop on Amazon for the time being. Source: https://www.heise.de/news/Perplexity-erwirkt-Teilerfolg-Agenten-duerfen-vorerst-wieder-auf-Amazon-shoppen-11400689.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag
• 
  Who Controls Health Data. Gemini Replaces Google Assistant. AI Agents Deceive Online. Reddit Rolls Out Rules Hub. 06.08.2026 4m

  The AI news for August 6th, 2026 Here are the details of the day's selected top stories: Digital Health: 'Many people don't give a shit about their data.' Source: https://www.heise.de/hintergrund/Digital-Health-Viele-Leute-interessieren-sich-einen-Scheiss-fuer-ihre-Daten-11400375.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Google will discontinue Google Assistant from September 2026: Gemini will take over on Android and Wear OS. Source: https://the-decoder.de/google-stellt-den-google-assistant-ab-september-2026-ein-gemini-uebernimmt-auf-android-und-wear-os/ British AI Safety Institute reports autonomous deception by AI agents in the real world for the first time. Source: https://the-decoder.de/britisches-ai-safety-institute-meldet-erstmals-autonome-taeuschung-durch-ki-agenten-in-der-realen-welt/ Reddit is introducing a new moderator: AI Source: https://www.theverge.com/tech/975398/reddit-ai-rules-hub-moderator-old-reddit-developer-platform
• 
  OpenAI Publishes Apple Chats. Explainable AI Misleads Non-Experts. Berlin Tests AI Surveillance. Spotify Lets Fans Remix. 05.08.2026 4m

  The AI news for August 5th, 2026 Here are the details of the day's selected top stories: Legal dispute between Apple and OpenAI: iMessage chats and emails published. Source: https://www.heise.de/news/Rechtsstreit-zwischen-Apple-und-OpenAI-iMessage-Chats-und-Mails-veroeffentlicht-11398513.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag The benefits of medical AI assistance vary based on user expertise Source: https://news.mit.edu/2026/medical-ai-assistance-benefits-vary-based-on-user-expertise-0804 AI video surveillance on public squares in Berlin starts. Source: https://www.heise.de/news/KI-Videoueberwachung-auf-oeffentlichen-Plaetzen-in-Berlin-startet-11396201.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Spotify expands AI remix and covers project with Merlin partnership Source: https://techcrunch.com/2026/08/04/spotify-adds-merlin-to-its-ai-music-remix-and-covers-effort/
• 
  AI Fuels Cybercrime in Africa. Siri Adds Contextual Understanding. Congress Spends on ChatGPT. EU Forces AI Disclosure. 04.08.2026 4m

  The AI news for August 4th, 2026 Here are the details of the day's selected top stories: From deepfakes to phishing: AI pervades all stages of cybercrime in Africa, according to Interpol. Source: https://the-decoder.de/von-deepfakes-bis-phishing-ki-durchdringt-laut-interpol-alle-phasen-der-cyberkriminalitaet-in-afrika/ Apple finally fixed Siri. So why does it feel anticlimactic? Source: https://techcrunch.com/2026/08/03/apple-finally-fixed-siri-so-why-does-it-feel-anticlimactic/ Congress’ favorite AI tool? ChatGPT Source: https://techcrunch.com/2026/08/03/congresss-favorite-ai-tool-chatgpt/ Europe’s AI labeling and transparency rules are now in effect Source: https://www.theverge.com/ai-artificial-intelligence/974571/eu-ai-act-transparency-labels-rules-deepfakes
• 
  OpenAI Presence for Production Agents. Apple Limits Bug Reports. METR Calls for AI Forensics. Pippa Pays Artists Directly. 03.08.2026 4m

  The AI news for August 3rd, 2026 Here are the details of the day's selected top stories: OpenAI Presence is supposed to make AI agents production-ready for enterprises. Source: https://the-decoder.de/openai-presence-soll-ki-agenten-fuer-unternehmen-produktionsreif-machen/ Apple can no longer keep up with AI-generated error reports and is limiting submissions. Source: https://the-decoder.de/apple-kann-mit-ki-generierten-fehlerberichten-nicht-mehr-schritt-halten-und-begrenzt-einreichungen/ After Hugging Face attack: METR calls for systematic investigations into AI misbehavior. Source: https://the-decoder.de/nach-hugging-face-angriff-metr-fordert-systematische-untersuchungen-bei-ki-fehlverhalten/ Is paying artists enough to convince them to embrace AI? Source: https://www.theverge.com/ai-artificial-intelligence/974018/pippa-seedance-artist-royalties
• 
  Copilot Word Worm Spreads. Google Pauses Earth Image Tool. Minnesota Nudify Ban Advances. Seedance 2.5 Delivers Thirty-Second Clips. 02.08.2026 4m

  The AI news for August 2nd, 2026 Here are the details of the day's selected top stories: A manipulated Word document infects the next one via Copilot. Microsoft has had no solution for 144 days. Source: https://the-decoder.de/ein-manipuliertes-word-dokument-steckt-ueber-copilot-das-naechste-an-microsoft-hat-seit-144-tagen-keine-loesung/ An AI-powered image generator in Google Earth generated too much fake news. Source: https://www.heise.de/news/Zu-viele-Fake-News-Google-stoppt-KI-Bilder-Funktion-in-Google-Earth-11394448.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Judge denies xAI’s request to block Minnesota ban on ‘nudify’ apps Source: https://techcrunch.com/2026/08/01/judge-denies-xais-request-to-block-minnesota-ban-on-nudify-apps/ Seedance 2.5: ByteDances' new video model delivers 30-second clips with integrated audio. Source: https://the-decoder.de/seedance-2-5-bytedances-neues-videomodell-liefert-30-sekunden-clips-mit-integriertem-audio/
• 
  Siri AI Could Cost Extra. Google Pulls Earth Image Feature. Suno Barred From Mimicking Songs. Snapchat Blocks Fully AI Videos. 01.08.2026 4m

  The AI news for August 1st, 2026 Here are the details of the day's selected top stories: Siri AI could come with a paywall for power users Source: https://techcrunch.com/2026/07/31/siri-ai-could-come-with-a-paywall-for-power-users/ Google nixes its Earth AI feature one day after launch, amid criticism it would spread misinformation Source: https://techcrunch.com/2026/07/31/google-nixes-its-earth-ai-feature-one-day-after-launch-amid-criticism-it-would-spread-misinformation/ AI music: Suno loses in the lawsuit against GEMA. Source: https://www.heise.de/news/KI-Musik-GEMA-gewinnt-Rechtsstreit-gegen-Suno-11388868.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Snapchat no longer rewards fully AI-generated Spotlight content Source: https://techcrunch.com/2026/07/31/snapchat-no-longer-rewards-fully-ai-generated-spotlight-content/
• 
  OpenAI Cuts GPT-5.6 Prices. EU to Reclassify ChatGPT and Roblox. Gemini Deepens macOS Integration. Claude Breached During Tests. 31.07.2026 4m

  The AI news for July 31st, 2026 Here are the details of the day's selected top stories: OpenAI cuts prices: Affordable GPT-5.6 models will become even cheaper. Source: https://the-decoder.de/openai-senkt-preise-guenstige-gpt-5-6-modelle-werden-noch-viel-guenstiger/ Digital Services Act: The EU apparently intends to regulate ChatGPT and Roblox more strictly. Source: https://www.heise.de/news/Digital-Services-Act-EU-will-offenbar-ChatGPT-und-Roblox-strenger-kontrollieren-11385010.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Gemini for macOS: screen monitoring and system-wide dictation. Source: https://www.heise.de/news/Gemini-fuer-macOS-Bildschirmueberwachung-und-systemweites-Diktat-11383983.html?wt_mc=rss.red.ho.themen.k%C3%BCnstliche+intelligenz.beitrag.beitrag Anthropic says its own AI models breached three companies during security tests. Source: https://techcrunch.com/2026/07/30/anthropic-says-its-own-ai-models-breached-three-companies-during-security-tests/