AI 新闻 · V1

Claude Fable 5.2 and Mythos 5.2

Payments Canada announces launch of Canada's Real-Time Rail (RTR) payment system

Nemotron-Personas-Finance-Japan: A Synthetic Dataset Capturing Japan's Demographics and Financial Context

ChatGPT Enterprise now available in AWS Marketplace

Google Gemini routes personal Drive files and emails to human annotators

Semi-transparent, colored cadmium telluride photovoltaic panels in agrivoltaics

Claude Fable 5.2 Scores 65 on Artificial Analysis, Highest So Far

Our Smallest, Most Efficient Model: Muse Ember

DeepX claims inference up to 100B parameters at less than 5 W

More than 400k Russians killed fighting in Ukraine war, leaked data suggests

Show HN: STE80 Humanizer – de-AI text by writing 80% of the way to ASD-STE100

A year after the 100% AI mandate nobody understands our code

OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. We push the boundaries of the capabilities of AI systems and seek to safely deploy them to the world through our products. AI is an extremely powerful tool that must be created with safety and human needs at its core, and to achieve our mission, we must encom

Motion, plus Pages, Keynote, and Numbers that are getting access to Apple’s AI chatbot. [...] For Logic Pro, Apple says the usage limit is “impacted by the complexity of conversations you have with the assistant.” Users finding themselves running out of access can find the option to buy an Intelligence Usage Pack in the Intelligence Features > Show Usage section of the app’s main menu. Currently,

A challenge for newly minted products in the AI world is to retain people post-launch. Users might get excited to try out a new product, and then move on to the new shiny thing within days or weeks. Icoz is cognizant of the fact and said that he expects people to churn, as they might not be interested in making apps at all. [...] Icoz eventually disseminated the solution to his friends and added m

J Meta is launching its “smallest and most efficient” AI model: Muse Ember. The new model is “designed for complex reasoning execution and the work agents repeat all day: tool calls, coding subtasks, routing, extraction, and long document parsing,” Meta says. The company claims Muse Ember delivers “more intelligence per dollar” than Anthropic’s Claude Haiku 5.5 or OpenAI’s GPT-6 Luna “at any re

or AI agents, for access? “The generic web is getting dumber and dumber,” Stubblebine says. If AI is the future, “it needs to be based on this layer of human intelligence.” And all that intelligence might be better off outside the confines of the Google algorithm. [...] For now, TK is available to existing Medium members. (There are apps for the web, Android, and iOS, and Stubblebine says a native

Ember is the fourth model released under Meta’s Muse brand, following the mid-weight Spark (released in April), Spark’s open source twin Glimmer (released in August), and the Wonder frontier model (released in October). The flagship consumer AI product, the Muse Agent, was released in September to unexpected success, with new features like Spaces and video chat rolling out earlier this month. Top

# Jensen Huang reportedly engaged a PR firm with a history of handling high-profile sex scandals a year ago. Nvidia CEO Jensen Huang hired Heller Inc., a crisis communications firm known for handling sex scandals, at least a year before publicly acknowledged an extramarital affair with a woman suspected of spying for China, Wired reported. Heller Inc. founder Risa Heller has advised prominent men

Six months before the Nvidia spy scandal emerged, the New York-based short seller firm Culper Research had already raised questions about Nvidia CEO Jensen Huang’s unusually close relationship with Alice Huang, an executive at Megaspeed, a Singapore-based company that buys Nvidia chips. The Nvidia CEO admitted to an extramarital affair over the weekend, and the company has now opened a formal inve

That’s the finding of AI security researchers at Apiiro, who discovered thousands of secrets, including database passwords, API tokens, and root logins, amid Claude Code settings files embedded in GitHub repositories. When a user chooses the “Yes, and don’t ask again” option to approve running a command, that choice gets stored as part of a rule in the local “.claude/settings.local.json” file on t

Meta announced on Monday that it’s expanding its AI-powered age estimation technology on Instagram worldwide to identify more teens and place them under its Teen Account protections, as it faces numerous lawsuits accusing its platforms of harming teens. The social media giant says it will use the AI technology to find people it suspects to be teens, even if they use an adult birthday. “We’ll conti

Online publishing platform Medium on Monday launched a new app, called TK, which aims to bring AI technology deeper into the writing process. The idea behind TK is to make it easier to turn ideas into written words by incorporating an AI assistant into the editor itself, allowing users to ask for help by typing “@TK” followed by their request or prompt. In Medium’s app, TK takes on a new role as i

Nvidia’s stock slid more than 4% on Monday, the first trading day after its board said it would investigate CEO Jensen Huang over his relationship with Alice Huang, an executive U.S. intelligence believes was spying for China. “I would definitely wager that there’ll be a filing,” Bartlett said in an interview, adding that he wouldn’t be surprised if a case had been filed by the end of day Monday.

The partnership will give Disney ‘early access’ to AI tech for its resorts, cruise line, ESPN graphics, and more.. The Walt Disney Company has teamed up with Adobe to develop proprietary “artist-led” generative AI models to support Disney’s production pipeline. The multiyear partnership was announced today at Adobe’s MAX conference, and it will see Adobe providing Disney with early access to its

Geothermal company Fervo couldn’t have picked a better time to deliver its first power plants. In its third quarter earnings report, released Monday, the company said it now has $11.9 billion worth of contracted revenue on the books, up from $7.2 billion last quarter. As hyperscalers scour the earth for electricity to power their AI data centers, Fervo is on track to complete a 396 megawatt power

Nvidia CEO Jensen Huang has never been quiet about his belief that the U.S. should continue selling AI chips to China, rather than shutting the country off from the technology. The Financial Times reported Friday on the relationship with Alice Huang, a former executive at Megaspeed, an Nvidia customer. Jensen Huang admitted to the affair and apologized in a memo to the company on Sunday. U.S. inte

Asked about AI executives’ warnings about their models’ safety on 60 Minutes, Trump said, “I think that the risk is not nearly as real as it– as they talk, and I think the risk can be managed, and it’s managed.” He claimed the DOJ is managing that risk, saying, “We had a situation where somebody was reported to us to be doing something bad with AI. Trump refused to name the person or company he wa

# Jensen Huang pulls out of ElevenLabs Summit amid Nvidia controversy. Nvidia CEO Jensen Huang has pulled out of the ElevenLabs Summit scheduled for this Wednesday in New York, ElevenLabs, the voice AI company, confirmed Monday. He has also dropped out of other speaking engagements, including the Anthropic CEO Forum. Nvidia is an investor in ElevenLabs. The withdrawals comes after Huang admitted t

# Anthropic launches new Claude Fable and Mythos 5.2 models. Claude Fable 5.2 is designed to take on big technical challenges, and it’s available today. Anthropic is announcing two new models today: Claude Fable 5.2 and Claude Mythos 5.2, which it says are designed to take on “daunting” technical challenges. The company claims the new models show improvements in coding, reasoning, and vision, and

AI’s power crunch has become so acute that Anthropic, a company that has generally avoided buying physical infrastructure, is spending $3.4 billion on natural gas turbines. The AI company, which is planning to go public this year, is buying Caterpillar generators that equipment supplier OnePWR had originally ordered, according to a Financial Times report. Historically, Anthropic has preferred to r

Nominal, a four-year-old, Austin-based startup that offers an AI platform purpose-built for engineering and testing complex hardware, like aircraft, nuclear reactors, and smart factories, has raised another big round of financing that triples its valuation. On Monday, it announced a $240 million Series C at a $3 billion valuation led by Lightspeed Venture Partners and co-led by Insight Partners. T

DeepMind workers invoke '1984' over contract clause that keeps AI concerns out of public view - The Cool Down

Concordance of Large Language Model Recommendations with Multidisciplinary - Bioengineer.org

Multimodal Language Models Fall Short in Autonomous Rural Skin Lesion Triage - Bioengineer.org

Why telcos can't afford to ignore decision APIs - Fierce Network

Study Reveals Fundamental Limits of Large Language Models in Spatial Reasoning - Bioengineer.org

Trump says his deportation policy was ‘too tough’ and more in ‘60 Minutes’ interview - Deseret News

Lawyer builds private AI server with eight used Nvidia V100 GPUs — up to 215 tokens per second to keep legal files offline, costing about $2,400 per four-GPU board - Tom's Hardware

Artful Intelligence: How Does Fiction Imagine AI? - American Academy of Arts and Sciences

Deerfield taps AWS to fuel healthcare AI ambitions - Fierce Healthcare

Google Escalates Nvidia Rivalry With Specialized Versions of AI Chips - Bloomberg.com

Democrats Demand Answers on Trump’s A.I. Policy Amid Spy Plot - The New York Times

U.S. Commerce Dept. Inquiry Into Megaspeed Was Halted After Nvidia Intervention - The New York Times

AI used to predict cavities - Medical Xpress