Nachrichten — Seite139

Microsoft introduced its Copilot Fall Release on Thursday, spotlighting a new AI chatbot persona named Mico. Designed as an expressive, customizable companion, Mico serves as the official successor to the classic Clippy. The optional avatar can change colors, animate, and even perform cartwheel‑like motions, aiming to make voice interactions feel more natural. A hidden Easter egg lets users tap the avatar until it temporarily transforms into Clippy, while a "real talk" feature promises conversational nuance. The announcement video highlights Mico at the 42‑minute mark, underscoring Microsoft's push toward a more personable AI experience.Weiterlesen

Researchers from Texas A&M, the University of Texas and Purdue University have introduced the “LLM brain rot hypothesis,” suggesting that continual pre‑training on low‑quality web text can cause lasting cognitive decline in large language models. Their pre‑print paper analyzes a HuggingFace dataset of 100 million tweets, separating “junk” tweets—identified by high engagement yet short length or superficial, click‑bait content—from higher‑quality samples. Early results show a 76 percent agreement between automated classifications and graduate‑student evaluations, highlighting the potential risks of indiscriminate data ingestion for AI systems.Weiterlesen

OpenAI announced it has acquired Software Applications, Inc., the developer of Sky—a desktop AI assistant for Mac that can see the screen and act within apps. The purchase brings together OpenAI's large language models with a product designed to work alongside users throughout their day, from writing to coding. Sky’s founders, Ari Weinstein and Conrad Kramer, previously sold Workflow to Apple, where it became Shortcuts, while co‑founder Kim Beverett spent a decade at Apple on Safari and other core services. The deal, led by OpenAI executives Nick Turley and Fidji Simo, adds a new layer to Apple’s emerging AI ecosystem, which includes Apple Intelligence and a partnership that routes Siri queries to ChatGPT.Weiterlesen

Microsoft has introduced Copilot Mode to its Edge browser, offering AI‑driven assistance such as Actions and Journeys. The feature, available in a limited preview for U.S. users, lets people talk to the browser, have it perform tasks, and revisit previous browsing sessions. Edge now joins competitors like OpenAI’s ChatGPT Atlas, Perplexity’s Comet, and Google’s Gemini in Chrome, intensifying the AI‑focused browser wars. Users can toggle the mode on or off, and Microsoft says data collection is limited to improving the experience while preserving privacy.Weiterlesen

Meta has rolled out AI‑powered photo and video editing features that users can access straight from Instagram Stories. By tapping the paintbrush icon and selecting the “Restyle” menu, creators can enter text prompts to add, remove, or change elements in their media, from hair color to background scenery. The rollout expands upon earlier AI capabilities that were limited to chatbot interactions, adding preset effects, style filters, and video overlays such as snow or flames. Meta also highlighted related AI updates, including a new AI‑generated video feed called “Vibes” and upcoming parental‑control options that let parents manage AI chat interactions for teen users.Weiterlesen

Electronic Arts announced a partnership with Stability AI, the creator of the Stable Diffusion model, to co‑develop AI models, tools, and workflows that will help artists, designers, and developers reimagine game content creation. EA emphasizes that AI will act as a trusted ally while keeping humans at the center of storytelling. Early projects include accelerating physically based rendering material creation and enabling AI‑driven pre‑visualization of 3D environments. The move reflects EA’s broader push to embed AI across its operations and aligns with industry trends toward AI‑first strategies.Weiterlesen

Turbo AI, founded by 20‑year‑old college dropouts Rudy Arora and Sarthak Dhawan, has grown from one million to five million users within six months. The Los Angeles‑based startup offers an AI‑driven note‑taking and study tool that records, transcribes, summarizes, and creates flashcards, quizzes, and a chat assistant. Its user base now includes students at top universities and professionals at firms such as Goldman Sachs and McKinsey. With eight‑figure annual recurring revenue, cash‑flow positivity, and a cautious $750,000 funding round, Turbo AI remains profitable while exploring new pricing models.Weiterlesen

A Deloitte survey of U.S. consumers shows that while more than half are experimenting with or regularly using generative AI, a majority also express strong worries about privacy and security. About four in ten respondents pay for AI services, yet concerns about data misuse, inaccurate results, and companies’ focus on competition over problem solving persist. Users increasingly verify AI outputs and remain reluctant to share personal data, highlighting a trust gap that tech firms must address.Weiterlesen

Over 700 high‑profile individuals, including AI pioneers and celebrities, have signed a statement demanding a halt to the creation of artificial superintelligence until it can be proven safe. The petition warns that unchecked AI could threaten freedom, national security, and even human survival. A recent poll shows a clear public desire for stricter regulation, with a majority opposing rapid, unregulated AI progress. The movement reflects growing unease about the pace of AI advances and calls for stronger oversight before further development.Weiterlesen

China has unveiled a draft five-year plan that emphasizes technological self‑reliance and reduced exposure to foreign pressure. The proposal highlights development in semiconductors, artificial intelligence and renewable energy, while seeking to boost domestic consumption and lessen reliance on exports. It builds on the previous plan launched during the earlier U.S. administration and reflects ongoing trade tensions between Beijing and Washington.Weiterlesen

Microsoft introduced a fresh set of capabilities for its Edge browser’s Copilot Mode, positioning it as an AI‑driven companion that can view and reason over open tabs, summarize content, and perform actions like filling forms or booking hotels. The rollout, which follows a similar launch of OpenAI’s Atlas browser, adds new “Actions” and “Journeys” features and builds on the earlier July debut of basic Copilot functionality. The move underscores a growing competition to embed artificial intelligence directly into web browsers.Weiterlesen

Microsoft is adding a visual, animated companion named Mico to its Copilot AI in Windows 11. Designed as an expressive, customizable blob with a face that reacts to user interactions, Mico aims to make the chatbot feel more human‑centered. The move recalls earlier Microsoft assistants such as Clippy, Microsoft Bob, and Rover, but ties the familiar visual style to modern large‑language models and voice controls. Microsoft hopes the new interface will overcome past criticisms of its digital assistants by providing richer context awareness and a friendlier user experience.Weiterlesen

OpenAI announced it has purchased Software Applications Incorporated, the creator of the Mac automation app Sky. Sky, which leverages agentic AI to understand screen content and act across applications, will be integrated into ChatGPT. The acquisition also brings the entire Software Applications team into OpenAI. The move follows Apple’s recent work on Siri enhancements and comes shortly after OpenAI released its first web browser, ChatGPT Atlas, and bought the personal‑investing app Roi.Weiterlesen

Reddit has filed a lawsuit in the U.S. District Court for the Southern District of New York against AI search developer Perplexity and three data‑scraping firms—Oxylabs UAB, AWMProxy and SerpApi—accusing them of illegally harvesting Reddit content and violating the platform’s copyright protections. The complaint alleges the firms bypassed technical barriers, accessed billions of search‑engine result pages, and traced the scraped data back to Perplexity, which had previously received a cease‑and‑desist letter. Reddit, which hosts over 110 million daily active users and more than 22 billion posts and comments, has previously licensed its data to OpenAI and Google and has taken legal action against other AI companies over similar concerns.Weiterlesen

Microsoft introduced Mico, a visual avatar for its Copilot AI chatbot, at the company’s Copilot fall release event. Designed as a warm, customizable "blob" that reacts to user interaction, Mico offers an expressive face for the AI, complete with color changes and an Easter‑egg that transforms it into the classic Clippy. Available initially in the United States, Canada, and the United Kingdom, Mico supports voice mode, long‑term memory, and a new "Real Talk" personality mode that mirrors users without being overly sycophantic. The rollout signals Microsoft’s push to humanize AI while expanding Copilot’s productivity features.Weiterlesen

Anthropic announced a major upgrade to its Claude chatbot, adding an enhanced memory feature set for paying users. The new capabilities let Claude remember past conversations, learn user preferences, and compartmentalize information through a Projects feature that creates separate memory spaces. Users can edit, import, or export Claude’s stored context, and the company emphasized transparency by showing a synthesis of what the model retains. Anthropic also reported extensive safety testing to curb sycophancy and harmful outputs. The optional feature is available now for Max subscribers, with Pro access slated for the near future.Weiterlesen

Microsoft introduced Mico, an expressive, customizable visual avatar designed to accompany the Copilot voice assistant. Positioned as a modern take on the classic Clippy, Mico appears as a responsive blob that changes color and expression based on user interaction. The AI companion aims to provide respectful, supportive guidance, even pushing back when needed. Integrated with Copilot’s new Learn Live feature, Mico serves as a Socratic tutor, guiding users through concepts with questions, visual cues, and an interactive whiteboard. Initially available in the United States, United Kingdom, and Canada, users can enable or disable the visual presence at will.Weiterlesen

Anthropic is adding a Memory feature to its Claude AI chatbot, now available to paid Pro and Max users. The tool lets Claude retain information from past conversations, creating a more personalized and efficient experience. Users can view, edit, and delete stored memories, and even import data from other AI platforms such as ChatGPT and Gemini. Distinct memory spaces allow separate projects to stay organized, while granular controls let users toggle memory on or off. The rollout follows recent enhancements like real‑time web search, Voice Mode, and integrations with Google Workspace, Microsoft 365, Canva, and Figma.Weiterlesen

Anthropic has rolled out a new memory feature set for its Claude AI chatbot, allowing the model to retain context across conversations and create distinct memory spaces for different projects. The upgrade is optional and initially available to paid subscribers, with broader rollout planned for Pro users. Claude’s memory is transparent, letting users view and edit stored information, and supports import/export with other AI platforms. Anthropic emphasized extensive testing to maintain safety and reduce sycophancy, positioning the update as a step toward more personalized and secure AI interactions.Weiterlesen