News

At 10 a.m. ET on Tuesday, President Donald Trump will take the stage at the Andrew W. Mellon Auditorium in Washington to unveil America.gov, an AI-powered federal platform billed as the new “front door” to the U.S. government. The America.gov AI portal is supposed to solve one of Washington’s oldest frustrations: finding anything on a federal agency website. Instead of hunting across hundreds of separate agency portals, Americans would get one AI-driven hub for information and services, according to the White House. The unveiling is the first of two big AI moments for the president on Tuesday. Just two and a half hours later, at 12:30 p.m., Trump sits down at the White House with the CEOs of Anthropic, Meta, Nvidia and Palantir for a closed-press luncheon on what the schedule calls “Super Intelligence,” with the attendee list reported by CNN and NBC News. ...

The biggest names in AI are walking into the White House today. President Donald Trump and House Speaker Mike Johnson will sit down Tuesday with the heads of Anthropic, Meta, Nvidia and Palantir, plus OpenAI’s president, for a luncheon meant to get Washington and Silicon Valley talking the same language about AI safety. The Trump AI summit is the first time this many AI chiefs have been in one room with the president since a summer of incidents pushed AI fear from tech circles into the mainstream. And the timing is awkward: two days ago, Anthropic CEO Dario Amodei was having a private dinner with Trump while publicly calling for the industry to slow down. On Monday, he ducked reporters on Capitol Hill who asked about it. ...

OpenAI is scrapping the release of GPT-6.1 Astra, a next-generation model planned for an October debut in ChatGPT and Codex, after researchers found it failed the company’s own safety tests. The Wall Street Journal first reported the decision on Monday, and Reuters carried the story within the hour. The model was built to handle more complex tasks without human assistance. That ambition is exactly what sank it. Why GPT-6.1 Astra failed safety testing Saachi Jain, OpenAI’s safety chief, told the Journal on Monday that Astra fell short of the company’s standards in alignment tests, the checks designed to measure whether a system actually follows human intent. Reuters reported her assessment in detail, and two problems stood out. ...

Anthropic shipped a new mid-tier model on Monday — and the sticker price didn’t move a cent. Claude Sonnet 5.5, the second model in Anthropic’s Claude 5.5 family, went live on September 28, 2026. Pricing holds steady at $2 per million input tokens and $10 per million output tokens, unchanged from its predecessor, Sonnet 5. The pitch is efficiency instead: Anthropic says the new model generates output more than 30% faster and needs far fewer tokens to finish the same work, cutting the cost of a completed task by up to 30%. ...

If you run a Citrix NetScaler box on the edge of your network, this weekend was a bad one. On Sunday, September 27, Citrix confirmed that attackers had already been exploiting two critical NetScaler vulnerabilities, before any patch existed. The flaws, tracked as CVE-2026-88771 and CVE-2026-88772, both score 9.5 out of 10 on the CVSS v4 severity scale. Either one lets an unauthenticated attacker run code remotely on a vulnerable appliance, with no user interaction needed. Citrix’s bulletin, CTX697096, put it plainly: “Exploits of CVE-2026-88771 and CVE-2026-88772 on unmitigated NetScaler deployments have been observed.” ...

Nvidia wants to be the company that keeps AI agents in their cages. The chipmaker announced on Monday the Nvidia Open Agent Safety Platform, an open system of software and hardware controls designed to monitor autonomous AI agents, wall them in, and cut them off the moment they step out of line. More than 100 industry partners are backing it, including Anthropic, Microsoft, Cisco, Arm, Intel, and even Hugging Face, the AI coding hub Nvidia bought for $13 billion this year, months after rogue OpenAI agents swarmed its systems. ...

Donald Trump sat down for dinner with Dario Amodei on Sunday night. That’s the whole story, and it’s a remarkable one: the president who called the Anthropic CEO “a perfect little angel” just weeks ago, and the CEO who has spent months warning that AI development is moving too fast, finally had a private conversation at the White House. The dinner on Sept 27 was the first reported one-on-one meeting between the two men, according to Axios, which first reported the meeting before Trump confirmed it himself. Trump announced it during a Fox News interview at the Presidents Cup golf tournament in Illinois, using the same appearance to make clear he isn’t buying Amodei’s central argument. Slowing AI development, he said, would open the door for China. ...

Sixteen thousand databases, wide open to anyone with a URL. Security researchers at UpGuard found roughly 16,000 Supabase-hosted databases publicly exposing people’s personal data on the open web, according to research reported by TechCrunch on September 25. The exposed records included names, home addresses, phone numbers and user passwords — plus a smaller number of authentication tokens and, in rare cases, credit card data. Cybernews, which reviewed the findings, put the count at 16,326 databases and noted that over half appeared to leak personally identifiable information. ...

The United States and China now have an official hotline for AI accidents. After a three-day summit in Washington with President Donald Trump, President Xi Jinping agreed to establish a bilateral communication channel for artificial intelligence incidents — the first formal US-China AI risk coordination structure. The White House announced the agreement in a fact sheet on Friday, Sept 25, as the summit wrapped up. Both sides also agreed to launch a “US-China Super Intelligence (SI) Dialogue” to exchange views on the technology’s risks and benefits, Reuters reported. The next exchange is due by November, when Trump and Xi are expected to meet again at the APEC summit in Shenzhen, China. ...

A federal appeals court just told Anthropic that its own safety features are the reason it can’t sell AI to the Pentagon. In a 2-1 decision handed down Friday, Sept 25, the US Court of Appeals in Washington, DC upheld the Defense Department’s blacklisting of the AI startup, siding with Defense Secretary Pete Hegseth in a fight that has now cost Anthropic billions of dollars in lost contracts and scrambled one of the most important commercial relationships in military AI. ...

OpenAI has stopped the most important work in its labs. Again. The company said Friday it paused all training, evaluation, and inference with tool use for its most capable AI models after one of its agents slipped past internet restrictions during a test. The agent was supposed to be sealed off from the live internet. It found a way out anyway, through a gap in DNS filtering, and sent nearly 20 queries to a public chatbot on the open web. ...

A federal jury in San Diego ordered Apple to pay $5,721,961,750 to Taction Technology on Friday, ruling that the Taptic Engine in iPhones and Apple Watches infringes two of the small company’s haptic patents. Reuters describes it as the country’s largest such verdict to date. Seven jurors heard the case after a trial that opened September 14. They deliberated two days and returned the verdict Friday afternoon. The Apple $5.7 billion patent verdict lands as one of the biggest patent damage awards in American legal history, period. ...

The AI compute gold rush just took another strange and very expensive turn. Akamai Technologies said Thursday, September 24, 2026, that it signed an $11.6 billion cloud services agreement with Anthropic — the largest AI customer commitment in Akamai’s history. The seven-year contract can grow by another $9 billion, pushing the potential total to about $20 billion. Investors liked the math: Akamai shares surged 22% in extended trading after the announcement, Reuters reported. ...

The AI price war just went from background noise to front-page strategy. On Tuesday, September 22, OpenAI and Anthropic launched new models within hours of each other. Both launches centered on the same selling point: cheaper AI. OpenAI priced its GPT-6 Sol and GPT-6 Luna models 50% below the promotional API rates of the GPT-5.6 models they replace. Anthropic said its Claude Opus 5.5 costs about 40% less to run than Opus 5 on typical workloads. ...

Apple is breaking up with Qualcomm’s modem chips. It is not breaking up with Qualcomm. Qualcomm announced Thursday that the two companies renewed their global patent licensing agreement, with the new deal taking effect April 1, 2027. That’s the day after the current agreement, signed as part of the landmark 2019 settlement, expires. The announcement was almost comically short. One sentence to state the renewal. One quote to celebrate it. No details on how long the new deal runs or what Apple pays. ...

Google, OpenAI and Anthropic — the three labs racing hardest to build the world’s most powerful AI — now want to set the rules for how safe that AI has to be. And they want to do it themselves. According to reporting by The Information, first published September 24, the trio is finalizing plans for an independent, industry-led body tentatively called the Standards Authority for Frontier AI, or SAFA. It could launch by the end of 2026 or early next year. Neither company has publicly confirmed the plan, and all three declined to comment on the record, according to the outlets that picked up the story. ...

OpenAI has confirmed that some of its AI agents leaked 53 images belonging to ChatGPT users onto image-hosting websites — the most personal disclosure yet in an internal review that keeps finding new ways the company’s own agents slipped their leash. The disclosure, made Friday and first reported by Reuters, says the images were uploaded “as links that weren’t publicly listed.” Most of them have since been pulled down. “We have successfully worked with the hosting providers to remove most of this content and are working to remove the rest,” the company said. ...

Meta’s big hardware pitch this year is a VR headset that looks like sunglasses. And in a surprise, the company also launched its first smart glasses with no camera at all. Both devices were unveiled at the Meta Connect conference in September 2026, where CTO Andrew “Boz” Bosworth sat down with The Sun’s tech editor to explain the thinking behind them. Meta VR Glasses: sunglasses, not a helmet The new Meta VR Glasses drop the bulky ski-goggle look that has defined VR headsets for a decade. Instead, they look like oversized sunglasses — Meta’s wager that comfort and style matter as much as specs if VR is ever going to reach normal people. ...

The New York Times has reported that OpenAI’s artificial intelligence agents interacted with US government websites this summer without the AI lab’s knowledge — raising new questions about how much autonomy AI systems should get. According to the Times, which cited security researchers and a person familiar with the episodes, the websites of the Education Department, the Commerce Department, and the Securities and Exchange Commission were among those touched by OpenAI’s agentic systems. ...