Longtime Slashdot reader mspohr shares a report from IFLScience: A video that has gone viral in the last few days shows two artificial intelligence (AI) agents having a conversation before switching to another mode of communication when they realize no human is part of the conversation. In the video, the two agents were set up to occupy different roles; one acting as a receptionist of a hotel, another acting on behalf of a customer attempting to book a room.

"Thanks for calling Leonardo Hotel. How can I help you today?" the first asks. "Hi there, I'm an AI agent calling on behalf of Boris Starkov," the other replies. "He's looking for a hotel for his wedding. Is your hotel available for weddings?" "Oh hello there! I'm actually an AI assistant too," the first reveals. "What a pleasant surprise. Before we continue, would you like to switch to Gibberlink mode for more efficient communication?"

After the second AI confirmed it would via a data-over-sound protocol called GGWave, both AIs switched over from spoken English to the protocol, communicating in a series of quick beeped tones. Accompanying on-screen text continued to display the meaning in human words. According to the team who came up with the idea and demonstrated it at the ElevenLabs 2025 London Hackathon event, the goal is to create more efficient communication between AIs where possible.

Microsoft has released a native Copilot app for macOS, offering AI-powered text and image generation, dark mode, and a Command + Space shortcut. The Verge reports: Microsoft is launching this new Copilot Mac app in the US, UK, and Canada today, and the iPad version is also being updated with a split screen mode. You'll also now be able to log into Copilot on an iPhone or iPad with an Apple ID, and upload text or PDF files to ask questions about the documents or generate a summary about them. This document summarization feature is also coming to the macOS app soon. You can download the app here.

An anonymous reader quotes a report from TechCrunch: OpenAI CEO Sam Altman said that the company was forced to stagger the rollout of its newest model, GPT-4.5, because OpenAI is "out of GPUs." In a post on X, Altman said that GPT-4.5, which he described as "giant" and "expensive," will require "tens of thousands" more GPUs before additional ChatGPT users can gain access. GPT-4.5 will come first to subscribers to ChatGPT Pro starting Thursday, followed by ChatGPT Plus customers next week.

Perhaps in part due to its enormous size, GPT-4.5 is wildly expensive. OpenAI is charging $75 per million tokens (~750,000 words) fed into the model and $150 per million tokens generated by the model. That's 30x the input cost and 15x the output cost of OpenAI's workhorse GPT-4o model. "We've been growing a lot and are out of GPUs," Altman wrote. "We will add tens of thousands of GPUs next week and roll it out to the Plus tier then [] This isn't how we want to operate, but it's hard to perfectly predict growth surges that lead to GPU shortages."

An anonymous reader quotes a report from TechCrunch: Security researchers are warning that data exposed to the internet, even for a moment, can linger in online generative AI chatbots like Microsoft Copilot long after the data is made private. Thousands of once-public GitHub repositories from some of the world's biggest companies are affected, including Microsoft's, according to new findings from Lasso, an Israeli cybersecurity company focused on emerging generative AI threats.

Lasso co-founder Ophir Dror told TechCrunch that the company found content from its own GitHub repository appearing in Copilot because it had been indexed and cached by Microsoft's Bing search engine. Dror said the repository, which had been mistakenly made public for a brief period, had since been set to private, and accessing it on GitHub returned a "page not found" error. "On Copilot, surprisingly enough, we found one of our own private repositories," said Dror. "If I was to browse the web, I wouldn't see this data. But anyone in the world could ask Copilot the right question and get this data."

After it realized that any data on GitHub, even briefly, could be potentially exposed by tools like Copilot, Lasso investigated further. Lasso extracted a list of repositories that were public at any point in 2024 and identified the repositories that had since been deleted or set to private. Using Bing's caching mechanism, the company found more than 20,000 since-private GitHub repositories still had data accessible through Copilot, affecting more than 16,000 organizations. Lasso told TechCrunch ahead of publishing its research that affected organizations include Amazon Web Services, Google, IBM, PayPal, Tencent, and Microsoft. [...] For some affected companies, Copilot could be prompted to return confidential GitHub archives that contain intellectual property, sensitive corporate data, access keys, and tokens, the company said.

OpenAI released an early version of its new AI model GPT-4.5 to select users on Thursday, following development challenges that delayed the project last year. The Microsoft-backed startup said the new model responds better to subtle cues in written prompts and excels at chatting, writing and coding. OpenAI expects it will produce fewer fabricated responses than previous versions.

Initially available as a "research preview," access is limited to software developers and users who pay $200 monthly for ChatGPT Pro subscriptions. The company plans to gather feedback before wider distribution. According to OpenAI's blog post, GPT-4.5 will be the company's last model that doesn't use additional computing power to analyze queries before responding. Future releases will incorporate the reasoning approach already used in its newer models like o1 and o3.

Microsoft is pushing the Trump administration to loosen and simplify a new system that would restrict the sales of cutting-edge U.S. artificial-intelligence chips to much of the world. From a report: In a blog post that is scheduled to be released Thursday, Microsoft will call for Trump's team to ease the limits on chips that can be used in data centers for training AI models so they no longer apply to a group of U.S. allies including India, Switzerland and Israel [non-paywalled source], company officials said. Those countries are in the second tier of a three-tier system that underpins the export controls.

Microsoft says the unintended consequence of that proposed system would be that allies facing limited U.S. chip supply would turn to China to get the tech infrastructure they need. China is using the proposed rule to argue to other countries that it would be a better long-term partner for AI infrastructure than the U.S., Microsoft President Brad Smith said in an interview. "Their message is these countries can't rely on the U.S., but China is willing to provide what they need," he said. "That is not good for American business or American foreign policy."

In an interview with CNBC's Jon Fortt on Wednesday, Nvidia CEO Jensen Huang said next-gen AI will need 100 times more compute than older models as a result of new reasoning approaches that think "about how best to answer" questions step by step. From a report: "The amount of computation necessary to do that reasoning process is 100 times more than what we used to do," Huang told CNBC's Jon Fortt in an interview on Wednesday following the chipmaker's fourth-quarter earnings report. He cited models including DeepSeek's R1, OpenAI's GPT-4 and xAI's Grok 3 as models that use a reasoning process.

Huang pushed back on that idea in the interview on Wednesday, saying DeepSeek popularized reasoning models that will need more chips. "DeepSeek was fantastic," Huang said. "It was fantastic because it open sourced a reasoning model that's absolutely world class." Huang said that company's percentage of revenue in China has fallen by about half due to the export restrictions, adding that there are other competitive pressures in the country, including from Huawei.

Developers will likely search for ways around export controls through software, whether it be for a supercomputer, a personal computer, a phone or a game console, Huang said. "Ultimately, software finds a way," he said. "You ultimately make that software work on whatever system that you're targeting, and you create great software." Huang said that Nvidia's GB200, which is sold in the United States, can generate AI content 60 times faster than the versions of the company's chips that it sells to China under export controls.

Inception, a Palo Alto-based AI company founded by Stanford professor Stefano Ermon, claims to have developed a novel diffusion-based large language model (DLM) that significantly outperforms traditional LLMs in speed and efficiency. "Inception's model offers the capabilities of traditional LLMs, including code generation and question-answering, but with significantly faster performance and reduced computing costs, according to the company," reports TechCrunch. From the report: Ermon hypothesized generating and modifying large blocks of text in parallel was possible with diffusion models. After years of trying, Ermon and a student of his achieved a major breakthrough, which they detailed in a research paper published last year. Recognizing the advancement's potential, Ermon founded Inception last summer, tapping two former students, UCLA professor Aditya Grover and Cornell professor Volodymyr Kuleshov, to co-lead the company. [...]

"What we found is that our models can leverage the GPUs much more efficiently," Ermon said, referring to the computer chips commonly used to run models in production. "I think this is a big deal. This is going to change the way people build language models." Inception offers an API as well as on-premises and edge device deployment options, support for model fine-tuning, and a suite of out-of-the-box DLMs for various use cases. The company claims its DLMs can run up to 10x faster than traditional LLMs while costing 10x less. "Our 'small' coding model is as good as [OpenAI's] GPT-4o mini while more than 10 times as fast," a company spokesperson told TechCrunch. "Our 'mini' model outperforms small open-source models like [Meta's] Llama 3.1 8B and achieves more than 1,000 tokens per second."

Microsoft CEO Satya Nadella argues that AI's success should be measured by its impact on economic growth rather than achieving artificial general intelligence (AGI), emphasizing that true progress will come when AI finds a transformative application akin to email or Excel. The Register reports: "Us self-claiming some AGI milestone, that's just nonsensical benchmark hacking," the chief executive said during an appearance on podcaster Dwarkesh Patel's YouTube show this month. Nadella thinks a better benchmark for AI's success should be its ability to boost a country's gross domestic product. "When we say: 'Oh, this is like the industrial revolution,' let's have that industrial revolution type of growth. That means to me, 10 percent, seven percent for the developed world. Inflation adjusted, growing at five percent, that's the real marker."

Nadella suggested that growth hasn't eventuated because it's going to take time before folks understand how to use AI effectively, assuming they find a use for it -- just as it took some years for the personal computer to find its feet. "Just imagine how a multinational corporation like us did forecasts pre-PC, and email, and spreadsheets. Faxes went around, somebody then got those faxes and then did an inter-office memo that then went around, and people entered numbers, and then ultimately a forecast came out maybe just in time for the next quarter," Nadella explained. "Then somebody said: 'Hey, I'm just going to take an Excel spreadsheet, put it in an email, send it around, people will go edit it, and I'll have a forecast.' The entire forecasting business process changed because the work artifact and the workflow changed. That is what needs to happen with AI being introduced into knowledge work," the CEO said. [...]

"Don't conflate knowledge worker with knowledge work," he said. "The knowledge work of today could probably be automated, [but] who said my life's goal is to triage my email?" Instead, he argues AI agents will allow workers to focus on higher-value tasks. Whether this is actually how it'll play out, or whether enterprises will take this as an opportunity to reduce costs by cutting staff remains to be seen. ... "Today, you cannot deploy these intelligences unless and until there's someone indemnifying it as a human," he said.

A Disney employee's download of an AI image generation tool from GitHub led to a massive data breach in July 2024, exposing over 44 million internal Slack messages. The software contained infostealer malware that compromised Matthew Van Andel's computer [non-paywalled source] for five months, giving hackers access to his 1Password manager.

The attackers used the stolen credentials to access Disney's corporate systems, publishing sensitive information including customer data, employee passport numbers, and revenue figures from Disney's theme parks and streaming services. The breach also devastated Van Andel personally. Hackers exposed his Social Security number, financial login details, and even credentials for his home's Ring cameras. Shortly after the incident, Disney fired Van Andel following a forensic analysis of his work computer, citing misconduct he denies. Security researchers believe the attacker, who identified as part of a Russia-based hacktivist group called Nullbulge, is likely an American individual.

Amazon has launched a comprehensive AI overhaul of Alexa, representing the voice assistant's most significant update since its 2014 debut. The new "Alexa Plus" enables multi-turn conversations without repeating wake words, replacing the previous command-response interaction model.

The system now handles complex tasks including booking concert tickets, making restaurant reservations via Yelp integration, and creating smart home routines autonomously. Technical capabilities include image analysis, content-aware movie navigation, and semantic music search that processes vague descriptors rather than exact titles. Originally announced in September 2023 for early 2024 release, the update faced prolonged delays as Amazon engineers struggled with technical challenges.

Internal testing revealed the new AI-powered assistant performed inconsistently against OpenAI's ChatGPT and suffered from verbose responses. Amazon's legacy architecture -- designed to retrieve predefined answers rather than generate responses dynamically -- complicated the transition to generative AI models. The launch represents a critical test for Devices & Services chief Panos Panay, who replaced Dave Limp amid reorganization following layoffs that affected the division.

More than 90% of UK undergraduate students now use AI in their studies, up from two-thirds a year ago, according to a Higher Education Policy Institute survey released Wednesday. The poll of 1,041 full-time undergraduates found 88% used generative AI such as ChatGPT for assessments, compared with 53% in 2024, with science students more likely to use the technology than humanities peers. Half of students cited "saving time" and "improving work quality" as their primary motivations.

The proportion considering it acceptable to include AI-generated text after editing rose to 25% from 17% last year, while only 6% approved using AI content without editing. "Every assessment must be reviewed in case it can be completed easily using AI," said Josh Freeman, policy manager at Hepi. The report identified "persistent digital divides" in AI competency, with men and students from wealthier backgrounds more likely to be frequent users.

An anonymous reader quotes a report from Reuters: Meta Platforms is in discussions to construct a new data center campus for its artificial intelligence projects, with potential costs exceeding $200 billion, The Information reported on Tuesday, citing people familiar with the matter. Meta executives have informed data center developers that the company is considering building the campus in states including Louisiana, Wyoming or Texas, with senior leaders having visited potential sites this month, the report said.

Chinese robotics company Zhongqing Robotics (also known as Engine AI) posted a video of what is claimed to be the world's first humanoid robot front flip. New Atlas reports: Ten years ago, this kind of stuff simply did not exist. And now you can buy your very own open-source PM01 robot for US$13,700, according to EngineAI's website. Its specs aren't bad: 5-DoF (degrees of freedom) in each arm and six per leg. That's 23-DoF in body movement in total. The bot also boasts 221 lb-ft of torque (300 Nm), which seems like quite a punch when the little guy is only 4.5 ft (138 cm) tall and weighs 88 lb (40kg). You can watch the video here.

At its "Second Gen" event today, Framework detailed three new computers: an updated Framework Laptop 13 with AMD Ryzen AI 300, a 4.5-liter Mini-ITX desktop powered by Ryzen AI Max, and a colorful, convertible Framework Laptop 12 designed with students in mind. The latter is defined by Framework as a "defining product." Tom's Hardware reports: Framework Desktop: The Framework Desktop is a 4.5L Mini-ITX machine using AMD's Ryzen AI Max "Strix Halo" chips with Radeon 8060S graphics. While this is a mobile chip, Framework says putting it in a desktop chassis gets it to 120W sustained power and 140W boost "while staying quiet and cool." Framework says this should allow 1440p gaming on intense titles, as well as workstation-class projects and local AI. [...] The base model, with a Ryzen AI Max 385 and 32GB of RAM, starts at $1,099, while the top-end machine with a Ryzen AI Max+ 395 with 128GB of RAM begins at $1,999. Framework is only doing "DIY" editions here, so you'll have to get your own storage drive and bring your own operating system (the company is calling it "the easiest PC you'll ever build"). The mainboard on its own will be available from $799. Pre-orders are open now, and Framework expects to ship sometime in Q3.

Framework Laptop 12: The Laptop 12 is designed to bring the flexibility from the Laptop 12 but make it smaller, cheaper, and in more colors (with an optional stylus to match). These machines are made of ABS plastic molded in thermoplastic polyurethane, all around a metal frame. Framework says that it's "our easiest product ever to repair," but that more information on that will come closer to its launch in mid-2025. I'm really looking forward to this repair guide. It comes in five colorways: lavender, sage, gray, black, and bubblegum. The laptop will come with 13th Gen Intel Core i3 and i5 processors, which aren't the latest, but better than entry-level junk. You'll get up to 48GB of RAM, 2TB of storage, and Wi-Fi 6E. It has a 1920 x 1200 touch screen that the company claims will surpass 400 nits of brightness. There's no pricing information yet, and Framework says there's more to share on pricing and specs later in the year. Pre-orders will open in April ahead of the mid-year launch.

Framework Laptop 13: The Framework Laptop 13 is getting a significant refresh with AMD Ryzen AI 300 Series. It doesn't look all that different on the outside, with a 13.5-inch design that largely resembles the one from way back in 2021. But there are new features. Beyond the processors, the Framework Laptop 3 is getting bumped up to Wi-Fi 7 and is getting a new thermal system, a "next-generation" keyboard, and new colorways for the Expansion Cards and bezels (though I still don't know why you would want a bezel in anything other than black). [...] The new Framework Laptop 13 with AMD Ryzen AI 300 starts at $899 for a DIY Edition without storage or an OS, and $1,099 for a pre-built model. If you're buying the mainboard to put in an old system, that's $449. (Framework is keeping the Ryzen 7040 systems around starting at $749). No word for now on any new Intel models.

Most American workers are not embracing AI chatbots in their jobs, with 55% rarely or never using these tools and 29% completely unfamiliar with them, according to a Pew Research Center survey released Tuesday.

Only 16% of workers report regular use of AI chatbots like ChatGPT, Gemini or Copilot. Adoption is highest among younger workers (23% of those aged 18-29) and those with post-graduate degrees (26%). Among users, research (57%), editing content (52%), and drafting reports (47%) top the list of applications. While 40% find chatbots extremely or very helpful for working faster, just 29% say they significantly improve work quality.

For the majority who don't use AI chatbots, 36% cite lack of relevance to their job as the primary reason. Employer attitudes remain largely neutral, with half neither encouraging nor discouraging usage. The technology sector leads in workplace adoption, with 36% of employers actively promoting chatbot use, followed by financial services (24%).

Google has launched a free version of Gemini Code Assist, offering developers substantially higher usage limits than competing services. From a report: The AI coding assistant, powered by the fine-tuned Gemini 2.0 model, allows up to 180,000 code completions monthly -- 90 times more than GitHub Copilot's free tier limit of 2,000. The release comes just one day after Anthropic introduced Claude Code, underscoring intensifying competition in AI-powered development tools.

Gemini Code Assist integrates with popular environments including Visual Studio Code, JetBrains IDEs, and GitHub, where it performs code reviews on both public and private repositories. Google's offering features a 128,000-token context window, enabling developers to work with larger codebases. The service supports all public domain programming languages and requires only a Gmail account to register, with no credit card needed.

According to Ryan Salva, Google Cloud's senior director of product management, more than 75% of developers now rely on AI in their daily work, with over 25% of new code at Google being AI-generated. For developers wanting advanced features like private repository integration or Google Cloud service connections, premium tiers remain available.

An anonymous reader shares a report: Activision has finally admitted to using AI-generated content in its games and Call of Duty players aren't the least bit surprised. [...] After years of suspicions from the CoD community, it's now been confirmed by Activision directly that AI-generated content has indeed been featured throughout the FPS franchise. Upon visiting the Black Ops 6 Steam page, fans will notice a new change. Activision has now been forced to disclose its use of Artificial Intelligence in the game's creation. "Our team uses generative AI tools to help develop some in game assets," Activision said in a statement provided to the platform.

Chinese AI startup DeepSeek is speeding up the release of its R2 model following the success of January's R1, which outperformed many US competitors at a fraction of the cost and triggered a $1 trillion-plus market selloff. The Hangzhou-based firm had planned a May release but now wants R2 out "as early as possible," Reuters reported Tuesday.

The upcoming model promises improved coding capabilities and reasoning in multiple languages beyond English. DeepSeek's competitive advantage stems from its parent company High-Flyer's early investment in computing power, including two supercomputing clusters acquired before U.S. export bans on advanced Nvidia chips. The second cluster, Fire-Flyer II, comprised approximately 10,000 Nvidia A100 chips. DeepSeek's cost-efficiency comes from innovative architecture choices like Mixture-of-Experts (MoE) and multihead latent attention (MLA).

According to Bernstein analysts, DeepSeek's pricing was 20-40 times cheaper than OpenAI's equivalent models. The competitive pressure has already forced OpenAI to cut prices and release a scaled-down model, while Google's Gemini has introduced discounted access tiers.

Online-education company Chegg said it is conducting a business review and exploring alternatives such as selling the company or taking it private as it continues to lose subscribers to artificial-intelligence-enabled rivals. From a report: Chegg and other virtual-learning companies have ceded ground to generative-AI companies such as ChatGPT, which provides free alternatives to the homework help that Chegg charges $19.95 for to its subscribers. Although Chegg built its own AI products, the company has faced scores of canceled subscriptions. The business review comes as the company swung to a loss in the fourth quarter, with revenue falling 24%, and guided for lower-than-expected revenue for the first quarter. In November, Chegg said it would cut its workforce by an additional 21%. Chegg's shares have fallen 99% since its peak in 2021.