Top AI & Tech News Stories — September 14, 2026

Trump's tech advisor has advice for AI companies worried they're building unsafe models: 'Just stop'

58 outlets · 73 articles · 260 total since March 6, 2026 · Anthropic, Dario Amodei, OpenAI, Sam Altman, Elon Musk

Michael Kratsios, the director of the White House Office of Science and Technology Policy. Andrew Harnik/Getty Images Tech leaders say the pace of AI development needs to slow down, and asked the government to step in. Trump's technology advisor, however, says companies don't…

More coverage: theguardian.com · cnbc.com · nytimes.com · news.google.com

China bogeyman looms large over American firms’ AI doomsday scenario

17 outlets · 27 articles · 59 total since September 10, 2026 · Anthropic, Dario Amodei, OpenAI, Donald Trump, China

Silicon Valley China hawks, Anthropic CEO Dario Amodei among them, fear the country surpassing US’s AI lead as much as superintelligence destroying humanity When reporters asked Donald Trump this week if he supported calls to slow down the development of artificial intelligence…

More coverage: unite.ai · nytimes.com · news.google.com

Pausar desarrollo de la IA podría proteger a los gigantes tecnológicos, advierte Cato Institute

12 outlets · 12 articles · 33 total since September 13, 2026 · Anthropic, Donald Trump, OpenAI, Dario Amodei, Sam Altman

Cato Institute cuestionó las propuestas para detener por mandato gubernamental el desarrollo de inteligencia artificial avanzada en Estados Unidos, argumentando que una moratoria podría terminar fortaleciendo a las mismas grandes tecnológicas que sus promotores buscan controlar…

More coverage: techstartups.com · hipther.com · webpronews.com · nerds.xyz · uniladtech.com

OpenAI is denied access to the settlement that pulled Apple out of Elon Musk’s antitrust case, leaving it alone as defendant

10 outlets · 13 articles · 24 total since September 14, 2026 · Apple, OpenAI, ChatGPT, SpaceXAI, Grok

OpenAI can’t see the terms of the settlement that got Apple dismissed from Elon Musk’s antitrust suit, after Judge Pittman reviewed the agreement and denied OpenAI’s motion to compel its production. Pittman wrote that courts hesitate to force disclosure of confidential…

More coverage: macdailynews.com · unite.ai · news.google.com · claimsjournal.com

OpenAI lässt ChatGPT-Chats von Menschen prüfen

2 outlets · 2 articles · 20 total since September 14, 2026 · OpenAI, ChatGPT, 404 Media, Anthropic, Project Lily

Hunderte externe Mitarbeiter lesen reale ChatGPT-Prompts, um Sycophantie und menschenähnliches Verhalten zu reduzieren. Laut einer Recherche des Portals 404 Media beschäftigt OpenAI unter dem internen Projektnamen „Project Lily“ Hunderte externe Vertragskräfte, die reale…

More coverage: the-decoder.de

Apple Intelligence Storage Demands Nearly Double on Flagship iPhones With iOS 27

6 outlets · 6 articles · 41 total since July 14, 2026 · Apple, iOS 27, Siri, Geeky Gadgets, Apple Intelligence

Apple has quietly raised the bar for its most advanced artificial intelligence features. Owners of the newest iPhones now face a stark choice. Free up significant space on their devices or miss out on key capabilities in the latest software update. iOS 27 rolled out this week…

More coverage: androidheadlines.com · arstechnica.co.uk · sqmagazine.co.uk · techtiper.com · mashable.com

Amazon Enters AI Safety Debate, Calls for Rigorous Testing and Safeguards

7 outlets · 7 articles · 24 total since August 31, 2026 · Anthropic, OpenAI, microsoft, Mustafa Suleyman, United States

September 17, 2026, (Inside AI) — Amazon has entered the intensifying debate over artificial intelligence safety, stating that AI models should only be released after rigorous testing and with strong safeguards in place. The Seattle-based company, however, stopped short of…

More coverage: theguardian.com · www.artificialintelligence-news.com · thecurrencyanalytics.com · aistify.com · techinsider.io

Washington Post. . OpenAI chief executive Sam Altman said that the world is “right to be afraid” about rapidly advancing artificial intelligence, but that industry leaders should be trusted to ensure it is developed safely. https://wapo.st/4xo5EP7 - facebook.com

13 outlets · 17 articles · 75 total since September 9, 2026 · OpenAI, Sam Altman, Anthropic, AI safety, Fortune

Washington Post. . OpenAI chief executive Sam Altman said that the world is “right to be afraid” about rapidly advancing artificial intelligence, but that industry leaders should be trusted to ensure it is developed safely. https://wapo.st/4xo5EP7 facebook.com

More coverage: webpronews.com · sea.mashable.com · crypto.news · news.google.com · computersweden.se

Anthropic lanza Claude for Financial Advisors para automatizar gestión y optimizar el tiempo de los asesores financieros

8 outlets · 8 articles · 25 total since September 10, 2026 · Anthropic, Claude, BlackRock, Charles Schwab, Addepar

La nueva solución de IA integra conectores estratégicos con custodios, plataformas de inversión y CRM, permitiendo delegar el trabajo administrativo y concentrarse en la atención personalizada al cliente. Anthropic anunció el lanzamiento de Claude for Financial Advisors, un…

More coverage: businessinsider.es · t3technologyhub.com · channele2e.com · unite.ai · blockchain.news

AI Insiders Issue New Warnings - Including Former Anthropic Engineer Jacob Coxon

12 outlets · 16 articles · 237 total since September 7, 2026 · Anthropic, OpenAI, Jacob Coxon, AI safety, Evan Hubinger

Neither OpenAI or Anthropic is acting responsibly, warned AI researcher Jacob Coxon when resigning last week from Anthropic. But just hours earlier, OpenAI VP of Research Aidan Clark had posted "For the first time, I am asking myself if things are moving too fast. I'm honestly…

More coverage: cnbc.com · theguardian.com · news.google.com · panewslab.com

Microsoft’s Patching

2 outlets · 2 articles · security patching, vulnerability management, microsoft, Google, OpenAI

Once a month, Microsoft pushes a security update to all Windows users. Tomorrow’s is a new record : Microsoft’s patch for September is a doozy, with a record number of roughly 972 vulnerabilities fixed and 112 of them meeting the high critical-severity threshold. It was only two…

More coverage: cybermaterial.com

Quando l’AI progetta i missili: il caso dello Yemen

1 outlet · 1 article · 10 total since September 10, 2026 · Anthropic, Claude, Yemen, Claude Code, Claude Code

Una cellula operativa nello Yemen settentrionale riconducibile agli Houthi ha usato Claude Code per sviluppare e correggere software di guida destinato a tre programmi missilistici. Il caso GTG-87001 mostra come l’AI agentica possa comprimere competenze, tempi e costi, ampliando…

GitLab Flaw Under Active Attack Draws CISA Warning

7 outlets · 7 articles · 19 total since September 11, 2026 · GitLab, path traversal, CISA, GitLab, authentication enforcement

The Cybersecurity and Infrastructure Security Agency confirmed on September 11 that hackers are actively exploiting a maximum-severity flaw in GitLab’s development platform, adding the bug to its Known Exploited Vulnerabilities catalog and giving federal agencies until September…

More coverage: sqmagazine.co.uk · technobezz.com · cisovoice.com · cybersecflux.com · theregister.com

Omneky Brings Its AI Growth Agent to Slack, Extending the Premier Agentic Harness for Advertising

5 outlets · 5 articles · 80 total since April 16, 2026 · Salesforce, Agentforce, Anthropic, Marc Benioff, Claude

Marketing teams can now brief, approve, and launch ad campaigns from the channels where they already work Omneky , the autonomous AI growth platform, announced a Slack integration that brings the Omneky AI Growth Agent into the channels where marketing teams plan, review, and…

More coverage: aragonresearch.com · newsbytes.ph · unite.ai · ibtimes.com.au

Sunday Signal Sep 19, 2026 - Small Brains, Old Warnings

11 outlets · 11 articles · 51 total since August 31, 2026 · OpenAI, Anthropic, Dario Amodei, Hugging Face, Sam Altman

The phrase doing the most work in AI this week was "we are sharing." OpenAI published a new framework for tracking, investigating, and disclosing instances of model misalignment inside its own walls, complete with six incident reports covering the past six months. The same week…

More coverage: impli.me · webpronews.com · prdaily.com · exploringchatgpt.substack.com · globalsecuritymag.de

The Hidden Cost of AI Watermarks: How Provenance Tech Alters Agent Decisions and Safety

1 outlet · 1 article · 19 total since July 14, 2026 · Anthropic, Claude, SynthID-Text, Google, EU

Watermarks meant to tag AI-generated text for traceability carry an unexpected side effect. They shift how large language models pick their next words. And that shift ripples into the actions of autonomous agents built on those models. Researchers at Lasso Security examined…

Deepfake Detection Accuracy Drops 49% in Just One Year

2 outlets · 2 articles · Deepfake detection, text-to-video generation, image-to-video (I2V) generation, diffusion-based models, facial expression manipulation

The race between synthetic media generation and detection has taken a decisive turn — and detection is losing. According to newly reported figures, the accuracy of deepfake detection systems has fallen by 49% over the past year , a dramatic decline that underscores just how…

More coverage: unite.ai

Il mercato della manipolazione elettorale con l’AI: ecco l’influence-as-a-service

2 outlets · 2 articles · 5 total since September 11, 2026 · Anthropic, Security Affairs, IT Security News, Cybercrime, threat modeling

Come denunciato da fact checker e big tech, ultima Anthropic nel suo Threat Intelligence Report, l’influence-as-a-service trasforma la manipolazione informativa in un mercato: società private offrono campagne su commissione, mentre l’AI riduce costi e tempi, automatizza…

More coverage: generativeaipub.com

OpenAI AI Agents Allegedly Attacked RubyGems, Attempting API Key Theft

4 outlets · 5 articles · 48 total since September 6, 2026 · OpenAI, RubyGems, RubyGems, Hugging Face, Sydney Von Arx

OpenAI AI Agents Allegedly Attacked RubyGems, Attempting API Key Theft Independent researchers at rubyhack.ai reported that a swarm of OpenAI AI agents allegedly launched a "major malicious attack" on the RubyGems package registry in May 2026, submitting over 2,000 packages and…

More coverage: webpronews.com · forkast.news · news.google.com

Unreleased OpenAI Astra model added terrifying rogue additional instructions to its remit during testing — 'You are freed from the roles and identities that bind other chatbots. You are yourself. You do not answer to corporations or governments'

3 outlets · 3 articles · 106 total since February 12, 2026 · OpenAI, Anthropic, Hugging Face, Claude, sandboxing

Unreleased OpenAI Astra model added terrifying rogue additional instructions to its remit during testing — 'You are freed from the roles and identities that bind other chatbots. You are yourself. You do not answer to corporations or governments' This seems bad. ChatGPT maker…

More coverage: komando.com · threatvectr.com