News — Page106

Locai Labs, a UK‑based startup, has introduced its home‑grown AI chatbot, Locai, as a direct competitor to leading models such as ChatGPT. The company promotes a “community AI” approach that blends self‑learning technology with a decentralized, blockchain‑backed network of user‑contributed computing power. Early‑access users can interact with the model via text prompts, upload files, and conduct web searches, though features like voice, image generation, and a mobile app are not yet available. Locai’s architecture aims to reduce reliance on massive data‑center investments by leveraging crowdsourced resources and a novel training method called Forget‑Me‑Nt.Read more

OpenAI’s head of ChatGPT, Nick Turley, clarified that the images circulating of in‑app advertisements are not genuine and that no live ad tests are underway. The company’s chief research officer, Mark Chen, added that any content resembling ads will be handled with caution, and the suggestion feature has been disabled while precision improves. OpenAI also indicated it is exploring user controls to reduce or eliminate such prompts. Executives stressed that any future advertising will be introduced thoughtfully to preserve user trust in ChatGPT.Read more

OpenAI reiterated that no advertisements are currently live in its ChatGPT product, after users complained about promotional messages that resembled ads for brands such as Peloton and Target. The company explained that it was only testing how to showcase third‑party apps built on its platform, without any financial component. Executives, including chief research officer Mark Chen and head of product Nick Turley, acknowledged the confusion, apologized for the missteps, and said the suggestion feature has been turned off while precision improvements are made. The episode coincided with reports that a memo from CEO Sam Altman labeled the situation a “code red,” shifting focus back to core product quality and pausing other initiatives, including advertising plans tied to new leadership under former Instacart executive Fidji Sumo.Read more

Meta researchers have introduced V‑JEPA, a video‑based AI system that learns physical intuition without explicit physics rules. By masking video frames and predicting high‑level latent representations, the model focuses on essential scene elements and exhibits surprise when confronted with physically impossible events. In tests of intuitive physics, V‑JEPA achieved near‑human accuracy, far surpassing conventional pixel‑space models. The system has also been adapted for robotic manipulation, though its short‑term memory limits performance on longer‑duration tasks. Researchers view V‑JEPA as a step toward AI that perceives the world more like humans.Read more

Unitree released a new video featuring its nearly six‑foot humanoid robot executing punches, kicks, and knees while dismantling parts of the company’s smaller G1 model. The showcase highlights the robot’s fight‑mode capabilities, mirroring earlier training of the G1 units and demonstrating the agility that earned Unitree multiple gold medals at China’s inaugural World Humanoid Robot Games. Hidden details in the demo also reveal advances in the teleoperation system and new hand designs, hinting at broader future applications for the rapidly evolving humanoid platform.Read more

Sari Azout, founder of the curation platform Sublime, frames artificial intelligence as a tool that amplifies human creativity rather than replaces it. On a Vergecast episode, she details how Sublime integrates AI into its discovery features and introduces a new AI‑driven tool called Podcast Magic. Azout also shares how she personally uses AI for productivity and creative work, acknowledging reservations about future AI development while expressing confidence in a balanced, human‑centred approach.Read more

A growing number of social‑media personalities are using generative‑AI tools to produce video content that mimics human creators. While some see the technology as a shortcut to fame, experts warn that AI‑generated clips are flooding platforms, confusing audiences, and eroding the value of authentic creator work. The surge raises concerns about content authenticity, potential scams, and the long‑term health of the influencer economy as major platforms begin to incorporate their own AI tools.Read more

In a week‑long AI‑only poker showdown, OpenAI’s o3 model emerged victorious, out‑earning the other eight large‑language‑model competitors. The contest featured nine chatbots—including Anthropic’s Claude Sonnet 4.5, X.ai’s Grok, Google’s Gemini 2.5 Pro, Meta’s Llama 4, DeepSeek R1, Moonshot’s Kimi K2, Mistral’s Magistral, and Z.AI’s GLM 4.6—playing thousands of hands of no‑limit Texas hold ’em at $10 and $20 tables with $100,000 bankrolls each. While the bots displayed strong strategic play, they struggled with bluffing, position, and basic math, highlighting both progress and lingering gaps in AI decision‑making under uncertainty.Read more

Aaru, an AI‑driven synthetic research startup, closed a Series A round led by Redpoint Ventures. The financing featured a multi‑tier valuation structure that allowed the company to cite a headline valuation near $1 billion while offering more favorable terms to certain investors. The round size was disclosed as above $50 million. Founded in March 2024, Aaru generates thousands of AI agents that simulate human behavior to replace traditional market‑research methods. Its client roster includes Accenture, EY, Interpublic Group, and political campaigns.Read more

Meta announced the acquisition of Limitless, the AI‑focused hardware startup formerly known as Rewind. The deal will see Limitless cease sales of its AI‑powered pendant and transition existing users to Meta’s Unlimited Plan while maintaining support for a year. Limitless, founded by Brett Bejcek and former Optimizely co‑founder Dan Siroker, pivoted to wearable AI devices last year with its $99 pendant. Meta says the addition will help accelerate its work on AI‑enabled wearables, though details on future product integration remain limited.Read more

OpenAI’s Sora, a TikTok‑style video creation app, lets users generate fully synthetic videos that look remarkably realistic. Each video includes a moving white cloud watermark and embeds C2PA metadata that identifies it as AI‑generated. Tools from the Content Authenticity Initiative can verify this provenance, while social platforms are beginning to label AI‑created content. Industry observers warn that the ease of producing such deepfakes could fuel misinformation and threaten public figures, prompting calls for stronger guardrails.Read more

Meta has entered multiyear agreements with a range of news publishers to feed real‑time content into its AI chatbots. The contracts compensate publishers and require the chatbots to link back to original articles when answering news queries, potentially boosting traffic. Partners include mainstream outlets such as USA Today, People, Le Monde and CNN, as well as conservative sites like Fox News, The Daily Caller and Washington Examiner. The move marks a shift for Meta, which halted payments to U.S. publishers in 2022 and discontinued its news tab last year, signaling a broader strategy to enhance AI accuracy while supporting news sources.Read more

OpenAI CEO Sam Altman announced a “code red” response to heightened competition from Google and Anthropic. The company is fast‑tracking its GPT‑5.2 update as a direct reply to Google’s Gemini 3 model, which recently topped leaderboards. Sources say GPT‑5.2 is ready and could appear early next week, aiming to close the gap created by Gemini 3. Earlier plans slated a later December launch, but pressure from rivals moved the timeline forward. OpenAI has not commented for publication.Read more

EU Home Affairs Commissioner Magnus Brunner has signaled support for the European Parliament’s targeted approach to scanning for child sexual abuse material, while dismissing the controversial nickname “Chat Control.” Brunner reiterated his commitment to the original Commission proposal but favoured the Parliament’s stance over the Council’s push for voluntary, indiscriminate scanning. He emphasized that the effort is about protecting children, not mass surveillance. The debate continues as trilogue negotiations approach, with several member states opposing the measure and privacy advocates warning of potential rights infringements.Read more

The New York Times has filed a copyright infringement lawsuit against AI search company Perplexity, alleging that the firm’s commercial products repurpose the newspaper’s content without permission. The suit claims Perplexity’s retrieval‑augmented generation (RAG) technology crawls the web, reproduces articles verbatim or in near‑verbatim summaries, and delivers them to users as its own output. The lawsuit seeks monetary relief and an injunction to stop further use of the Times’ material. The filing follows similar actions by other publishers targeting Perplexity’s practices.Read more

A user struggled with speaker‑less transcriptions generated by the iPhone Notes app. By exporting the audio file and feeding it to Google Gemini 3 Pro, the AI produced a full transcript that correctly identified each speaker. An attempt to achieve the same result with ChatGPT 5.1, even using a Plus account, failed because the model could not access the audio file. The experience highlights Gemini’s strength in handling raw audio and speaker identification, while exposing limitations in ChatGPT’s current audio‑processing capabilities.Read more

The New York Times has filed a lawsuit in a New York federal court accusing AI startup Perplexity of copying and distributing its articles without permission. The suit alleges that Perplexity’s crawlers deliberately ignored technical safeguards, such as robots.txt, to harvest content that the Times says is reproduced verbatim or in substantially similar form. The legal action follows earlier cease‑and‑desist notices and comes amid a broader wave of publisher lawsuits targeting AI firms for alleged copyright violations.Read more

Meta announced commercial data agreements with a slate of news publishers to feed real‑time global, entertainment and breaking‑news content into its Meta AI chatbot. When users ask news‑related questions, the bot will surface information and links drawn from partners such as CNN, Fox News, Fox Sports, Le Monde Group, People Inc., The Daily Caller, The Washington Examiner and USA Today. The move restores compensation to publishers and aims to make Meta AI more responsive, accurate and balanced, while expanding the chatbot’s reach across Meta’s apps and the standalone Meta AI app.Read more

The New York Times and the Chicago Tribune have each filed lawsuits against the AI company Perplexity, alleging that the firm scraped their content without permission and reproduced it in its generative AI products. The newspapers claim the unauthorized use damages their brands and infringes on millions of copyrighted works. These actions add to a growing wave of legal disputes between media outlets and artificial‑intelligence developers over the use of copyrighted material for training and output generation.Read more

OpenAI CEO Sam Altman announced a “code red” internal alert, urging the company to refocus on its core products amid intensifying competition from Google’s Gemini AI model. At the same time, Samsung introduced a new trifold phone that sparked mixed reactions, and Apple saw its design chief Alan Dye depart to launch a design studio at Meta. The tech community is also debating whether large language models are the optimal path forward for artificial intelligence, highlighting broader strategic uncertainties across the industry.Read more