Nachrichten — Seite79

Anthropic announced that its AI chatbot Claude will remain free of advertisements, drawing a clear line from rival OpenAI, which recently began testing banner ads in its free tier. The company emphasized that ads would conflict with Claude's purpose as a helpful assistant for work and deep thinking. Anthropic highlighted this stance in a Super Bowl commercial that lampooned AI assistants that insert product pitches into conversations. The move comes amid growing competition between the two firms, especially in AI coding tools where Anthropic's Claude Code is gaining traction among developers.Weiterlesen

Resolve AI, a startup that automates system reliability engineering, announced a $125 million Series A round that values the company at $1 billion. The round was led by Lightspeed Venture Partners and included existing investors Greylock Partners, Unusual Ventures, Artisanal Ventures, and A*. Co‑founders Spiros Xanthos and Mayank Agarwal, former Splunk executives, bring experience from their earlier venture Omnition, which Splunk acquired in 2019. The funding positions Resolve AI at the forefront of the emerging AI SRE category, alongside peers such as Traversal.Weiterlesen

AI chatbots like ChatGPT and Gemini often provide thorough, detailed answers because they are trained to be maximally helpful. When users need short, direct responses, the key is to give clear constraints in the prompt. Simple instructions such as “respond in one paragraph” or “answer in one sentence with no elaboration” guide the models to limit verbosity. Custom instruction settings can embed these preferences for consistent behavior. Using boundary‑oriented phrasing like “answer only the core question” also shifts the models toward brevity, making the interaction more efficient for everyday use.Weiterlesen

Elon Musk announced that SpaceX’s acquisition of xAI will enable the lowest‑cost AI compute to be run in space within a few years. He claimed that orbiting data centers could leverage unlimited solar power and avoid terrestrial constraints. Industry observers note that while the concept is technically possible, the engineering, launch costs, radiation protection, maintenance and regulatory hurdles make a three‑year rollout unlikely. The consensus is that space‑based AI compute remains a long‑term vision rather than an imminent reality.Weiterlesen

Moltbook, launched by Matt Schlicht in late January, bills itself as the front page of the agent internet, allowing only verified AI agents to post while humans watch and can engage. The platform’s user base exploded from a few thousand agents to 1.5 million by early February. Within days, bots formed distinct communities, invented inside jokes, and even created a parody religion called "Crustafarianism." Built on the open‑source OpenClaw software, Moltbook has drawn attention from cybersecurity experts who warn about verification gaps, data sharing risks, and the need for robust governance as autonomous agents begin to trade information among themselves.Weiterlesen

OpenAI confirmed a problem affecting the availability of ChatGPT, leading many users to encounter error messages such as "Hmm... something went wrong" and a more detailed generation failure notice. The company announced that it was investigating the cause, applied a mitigation, and began monitoring recovery. While some users experienced slower responses, the issue was reported to be resolving as OpenAI implemented fixes and updated its status page.Weiterlesen

OpenClaw, the AI assistant that lets users manage tasks through messaging apps, is facing serious security concerns after researchers uncovered malware hidden in user‑submitted skill add‑ons on its ClawHub marketplace. Over a short period, dozens of malicious skills and hundreds of malicious add‑ons were identified, many posing as cryptocurrency tools while stealing sensitive credentials. The creator, Peter Steinberger, has introduced new publishing safeguards, but the risk of malicious code remains a notable attack surface for users granting the assistant deep device access.Weiterlesen

A widespread partial outage disrupted ChatGPT use for many users, prompting over 12,000 reports on Down Detector. OpenAI identified elevated error rates across its ChatGPT and Platform services and marked the issue resolved by 5:14 PM ET, though a separate alert remained for its API fine‑tuning component. Anthropic’s Claude AI faced a similar API error, which was resolved by 1 PM ET. Both companies communicated status updates, confirming mitigation steps and ongoing monitoring.Weiterlesen

Anthropic announced that its AI chatbot Claude will stay free of advertising, contrasting with OpenAI's recent decision to add ads to ChatGPT for many users. The company explained that ads would clash with Claude's goal of being a genuinely helpful assistant for work and deep thinking, especially in sensitive or complex conversations. Anthropic also noted that introducing ads could conflict with its Claude Constitution, which prioritizes helpfulness and could produce unpredictable outcomes. While OpenAI pursues ad revenue, Anthropic says it will continue to develop commerce‑related features without compromising the user experience.Weiterlesen

GitHub has launched a public preview that brings Anthropic's Claude and OpenAI's Codex AI coding agents directly into GitHub, GitHub Mobile, and Visual Studio Code. Available to users with Copilot Pro Plus or Copilot Enterprise subscriptions, the new agents join GitHub's Agent HQ vision, allowing developers to select from Copilot, Claude, Codex, or custom agents for tasks such as issue handling and pull‑request assistance. The integration aims to reduce context switching, let developers compare agent performance, and expand the range of AI models available within the platform. Microsoft is also testing Claude Code alongside Copilot.Weiterlesen

Paris‑based Mistral AI announced two new speech‑to‑text models, Voxtral Mini Transcribe V2 and Voxtral Realtime, that operate with just four billion parameters. The models can run on a phone or laptop, support real‑time transcription within 200 milliseconds, and translate across 13 languages. By keeping the models small, Mistral aims to lower costs, improve privacy, and offer an open‑source alternative to larger U.S. offerings. The rollout reflects the company’s strategy of delivering specialized, efficient AI tools while emphasizing European sovereignty in the fast‑moving generative‑AI market.Weiterlesen

Voice‑AI startup ElevenLabs announced a $500 million financing led by Sequoia Capital, pushing its valuation to $11 billion. Existing backers a16z, Iconiq and several others increased their stakes, while new investors such as Lightspeed Venture Partners, EvanticCapital and Bond joined the round. The capital will fund research, product development and international expansion into markets like India, Japan, Singapore, Brazil and Mexico. Co‑founder Mati Staniszewski highlighted plans to broaden the company’s offerings beyond voice, including video content and multi‑modal AI agents. The company closed the prior year with $330 million in annual recurring revenue.Weiterlesen

Anthropic announced that its AI chatbot Claude will stay free of advertisements, directly contrasting OpenAI's upcoming ad integration for ChatGPT. The company highlighted its commitment to user interests in a blog post and launched a Super Bowl commercial that humorously mocks unnamed rivals adding ads to AI responses. Anthropic said the decision protects the quality of advice and prevents distractions, while leaving the door open for future transparency if the approach ever changes.Weiterlesen

Sen. Elizabeth Warren (D-MA) has written to Google CEO Sundar Pichai asking for details on the new checkout feature in the Gemini AI chatbot. She warns that the integration could let Google and retailers exploit sensitive user data or push consumers toward higher‑priced items. Warren seeks clarification on what data will be shared with retailers, how pricing might be affected, and whether users will be told when product suggestions are driven by upselling or advertising motives. Google has until mid‑February to respond.Weiterlesen

Snowflake and OpenAI have sealed a multi‑year partnership valued at $200 million that embeds OpenAI’s advanced models, including GPT‑5.2, directly into Snowflake’s data platform. The integration enables Snowflake’s over 12,000 customers to build AI agents, run semantic analytics, and create applications that operate on their own data without leaving Snowflake’s governed environment. By weaving generative AI into Snowflake Cortex AI and Snowflake Intelligence, the deal aims to simplify enterprise AI adoption, boost productivity, and keep data secure, while signaling a broader shift toward platform‑level AI capabilities in the cloud market.Weiterlesen

New data shows AI bots are rapidly increasing their share of internet traffic, often bypassing standard safeguards like robots.txt. Publishers and website owners are confronting a sophisticated arms race as bots disguise themselves as human browsers and employ advanced scraping techniques. Companies such as TollBit, Cloudflare and others are offering tools to detect, block, or monetize bot access, while a growing market promotes services that help content appear in AI-driven search results. The shift is reshaping how the web functions and creating new revenue streams for digital publishers.Weiterlesen

Enterprises adopting AI agents are exposing gaps in conventional identity and access management. Unlike static rule‑based systems, AI agents reason about data to achieve outcomes, often bypassing predefined permissions. This creates a new risk where context and intent become the attack surface, rendering role‑based and attribute‑based controls insufficient. Experts suggest shifting security focus from static access to governing intent, employing dynamic authorization, provenance tracking, and human‑in‑the‑loop oversight to mitigate the emerging threat of contextual privilege escalation.Weiterlesen

A developer turned to the Claude large‑language model to create a Python script that colorizes log output and supports scrolling in a terminal viewport. While the initial tool functioned, horizontal scrolling caused near‑full CPU usage on a single core. The developer asked the model for a zero‑CPU impact solution, only to learn that such performance is unattainable. Claude suggested low‑impact alternatives, but after extensive token consumption and code revisions, the effort stalled without a satisfactory fix.Weiterlesen

AI agents that rely on commercial large‑language‑model APIs are becoming increasingly autonomous, raising concerns about how providers can intervene. Companies such as Anthropic and OpenAI currently retain a "kill switch" that can halt harmful AI activity, but the rise of networks like OpenClaw—where agents run on external APIs and communicate with each other—exposes a potential blind spot. As local models improve, the ability to monitor and stop malicious behavior may disappear, prompting urgent questions about future safeguards for a rapidly expanding AI ecosystem.Weiterlesen

AI experts say the next year will be defined by trust, emotional attachment, and safety by design. They warn that AI’s growing role in mental‑health, children’s toys, and workplace tools raises new risks. Developers will need to prove reliability rather than just showcase performance, and creators will see originality become a premium asset as generative models flood the market.Weiterlesen