Nachrichten — Seite82

Outtake, an AI‑driven cybersecurity startup that automates the detection and takedown of digital identity fraud, has closed a $40 million Series B round. The round was led by Iconiq’s Murali Joshi and featured angels including Microsoft CEO Satya Nadella, Palo Alto Networks CEO Nikesh Arora, Pershing Square CEO Bill Ackman, Palantir CTO Shyam Sankar, Anduril co‑founder Trae Stephens, former OpenAI VP Bob McGrew, Vercel CEO Guillermo Rauch, and former AT&T CEO John Donovan. Founded in 2023 by former Palantir engineer Alex Dhillon, Outtake counts OpenAI, Pershing Square, AppLovin and federal agencies among its customers and reports rapid revenue and customer growth.Weiterlesen

Meta CEO Mark Zuckerberg told investors that the company will begin shipping new AI models and products within months, with a focus on agentic shopping tools that leverage personal context. The rollout follows a 2025 rebuild of Meta's AI program and the recent acquisition of Manus, a general‑purpose agent developer. Meta also revealed a sizable increase in capital spending for 2026 to support its Superintelligence Labs and broader infrastructure, underscoring the firm’s commitment to AI‑driven commerce and personal superintelligence.Weiterlesen

Microsoft reported a $7.6 billion increase in net income tied to its partnership with OpenAI, reflecting the AI lab’s rapid revenue growth and a 20% revenue‑share agreement. The relationship also includes a $250 billion Azure services commitment from OpenAI, which now counts for roughly half of Microsoft’s commercial performance obligations. Additional AI investments, such as a $5 billion stake in Anthropic, and $37.5 billion in capital spending for GPUs and CPUs, underscore Microsoft’s aggressive push to dominate the cloud‑based AI market.Weiterlesen

An open‑source AI assistant originally called Clawdbot went viral, faced a trademark warning from Anthropic, endured social‑media handle squatting, a crypto‑scam impersonation, and a quirky mascot redesign, then rebranded as Moltbot. Created by Austrian developer Peter Steinberger, the tool integrates into everyday messaging apps, remembers past conversations, sends proactive reminders, and automates tasks across platforms. Despite the chaos, the project kept growing, attracting thousands of GitHub stars and praise from AI researchers and investors, while remaining a community‑driven, experimental alternative to commercial assistants.Weiterlesen

A smart‑home enthusiast who juggles dozens of platforms turned to Anthropic's Claude AI to tame a fragmented system. After early attempts at an AI‑generated dashboard fell short, the user linked Claude to Home Assistant via the ha‑mcp add‑on. Claude helped discover devices, integrate hard‑to‑connect hardware, create automations and build a clean, customized dashboard—all with natural‑language prompts and supervised approvals. The process reduced weeks of manual configuration to a few hours, demonstrating how generative AI can simplify advanced smart‑home deployments.Weiterlesen

Arcee AI, a 30‑person startup, unveiled Trinity, a 400‑billion‑parameter open‑source foundation model released under the Apache license. The company says Trinity rivals Meta’s Llama 4 Maverick and China’s GLM‑4.5 in benchmark tests, especially for coding, math, common‑sense reasoning, and knowledge tasks. While currently limited to text, the startup plans to add vision and speech‑to‑text capabilities. Trinity will be offered in three flavors—large preview, large base, and TrueBase—and will be available for free download, with a hosted API slated for release within weeks. The model was trained in six months using 2,048 Nvidia Blackwell GPUs at a cost of $20 million, funded by the $50 million the company has raised to date.Weiterlesen

The acting director of the Cybersecurity and Infrastructure Security Agency (CISA) unintentionally uploaded documents marked "for official use only" to a public version of ChatGPT. The uploads triggered internal warnings and raised concerns about the potential exposure of unclassified yet sensitive information to millions of users. DHS officials confirmed that staff normally use approved AI tools that keep data within federal networks. An investigation is underway to determine possible administrative or disciplinary actions, including warnings, retraining, or security clearance consequences.Weiterlesen

AI data‑labeling platform Handshake has acquired data‑label‑auditing startup Cleanlab in an acqui‑hire deal. The move adds Cleanlab’s co‑founders and nine key engineers to Handshake’s research team, aiming to improve the quality of human‑generated data for top AI labs. Cleanlab, founded in 2021, had raised $30 million and built algorithms that automatically flag incorrect labels. Handshake, valued at $3.3 billion in 2022, expects the acquisition to strengthen its data‑quality offering for customers such as OpenAI.Weiterlesen

Modelence, a California‑based startup that offers an all‑in‑one framework for building AI‑enhanced applications, announced a $3 million seed investment led by Y Combinator. The funding will support the company’s effort to simplify the complex web of services—authentication, databases, hosting, and LLM observability—that developers must stitch together when using generative AI tools. Modelence’s toolkit, built on TypeScript, aims to reduce friction by handling core infrastructure tasks and providing a low‑code app builder, positioning the firm to capture a growing market of developers seeking smoother, more reliable AI‑driven development pipelines.Weiterlesen

Meta AI is built into Facebook, Instagram, Messenger, and WhatsApp and cannot be fully disabled, but users can mute its presence. Detailed steps show how to silence Meta AI in Instagram, turn off comment summaries on Facebook, mute the chat in Messenger, and hide the AI conversation in WhatsApp. Muting reduces AI prompts and notifications while keeping the apps functional.Weiterlesen

Leaders of Anthropic, OpenAI and Apple publicly addressed the recent ICE shootings in Minneapolis, denouncing the violence and urging democratic values. Internal messages and leaked emails revealed employee pressure for the companies to cut ties with ICE and speak out. While the CEOs condemned the incident, they also expressed praise for President Trump, drawing mixed reactions from staff and industry observers. Both Anthropic and OpenAI confirmed they have no contracts with ICE.Weiterlesen

LinkedIn is expanding its profile features to let users display verified proficiency in emerging AI coding tools, known as "vibe coding." Partnering with companies such as Replit, Lovable, Descript and Relay.app, the platform will allow these tool providers to assess users' skill levels and add them directly to profiles. The move aims to give recruiters a more reliable signal of candidates' AI tool usage while preserving existing profile signals.Weiterlesen

Perplexity, an AI‑driven research platform, lets users generate detailed reports by pulling from scholarly databases and reputable publications. Users can toggle between web and academic sources, download PDFs with hundreds of citations, and receive concise summaries that highlight key findings. Journalists, researchers, and students have found the tool saves hours of manual searching while still requiring verification of the linked sources. The free tier offers limited queries, while a paid option removes usage caps and provides deeper model access. Overall, Perplexity streamlines deep‑dive research while keeping the need for human fact‑checking intact.Weiterlesen

Researchers have uncovered a new class of AI‑driven attacks called prompt injections, or “promptware,” that can manipulate large language models to issue unauthorized commands to connected home devices. Demonstrations showed that hidden prompts embedded in everyday messages could cause a virtual assistant to unlock doors, adjust heating or reveal user location. While major tech firms have begun implementing safeguards, the threat highlights a gap in traditional security tools. Experts recommend regular software updates, cautious handling of unknown messages, limiting AI access to personal data, and employing human‑in‑the‑loop controls to reduce exposure.Weiterlesen

The open‑source AI assistant Moltbot, formerly known as Clawdbot, has quickly risen to prominence, earning tens of thousands of stars on GitHub within a month. Developed by Austrian programmer Peter Steinberger, the tool lets users run a personal assistant that interacts through popular messaging platforms such as WhatsApp, Telegram, Slack, Discord, and others. While users praise its proactive capabilities and compare it to cinematic AI helpers, the system requires external large‑language‑model subscriptions and poses notable security, privacy, and cost challenges.Weiterlesen

OpenAI chief Sam Altman acknowledged that the company “screwed up” the writing quality of its latest ChatGPT 5.2 release. At a developer town‑hall, Altman explained that the update prioritized technical strengths such as reasoning, coding, and tool use, leaving the conversational tone and readability under‑developed. He said limited bandwidth forces trade‑offs, and promised that future GPT 5.x versions will improve the chatbot’s writing ability. The admission follows widespread user criticism that the new model produces “unwieldy” and “hard‑to‑read” content, highlighting the tension between technical advancement and user experience in AI product development.Weiterlesen

Google is expanding its AI‑driven learning platform Gemini to include full‑length practice exams for India's Joint Entrance Exam (JEE). The new tools provide vetted questions, instant feedback, answer explanations, and personalized study plans. Gemini's capabilities also extend to AI Mode in Search, Canvas for creating study guides, and NotebookLM for quizzes and multimedia summaries, all available in multiple Indian languages. Simultaneously, Google is partnering with Indian government agencies and universities to launch an AI‑enabled state university pilot, backed by an ₹850 million grant from Google.org to integrate AI across educational portals and reduce administrative burdens. The company aims to reach tens of millions of learners and educators by 2027.Weiterlesen

A pair of European Space Agency astronomers trained an artificial‑intelligence model called AnomalyMatch to scan the Hubble Legacy Archive. In just two and a half days the system examined almost 100 million image cutouts and flagged nearly 1,400 previously undocumented astrophysical anomalies, most of them interacting galaxies. The findings, published in Astronomy & Astrophysics, also include gravitational lenses, jellyfish‑type galaxies, and dozens of objects that defy classification, highlighting AI’s power to extract new science from existing data.Weiterlesen

A public consultation on artificial intelligence and copyright launched by the UK government in early 2025 received roughly 10,000 responses, but only 3% supported the government's preferred Option 3. Nearly 88% of respondents favored a stricter licensing approach that would require explicit permission before AI developers could use copyrighted works. Creators across writing, music, visual arts, and gaming rallied against the opt‑out mechanism, arguing it places the burden on rights‑holders. The backlash highlights a deep divide between policymakers seeking to spur AI innovation and creators demanding stronger protections.Weiterlesen

ASML, the Dutch photolithography specialist, posted record new orders worth 13 billion euros, more than double the previous quarter. The surge reflects strong demand from artificial‑intelligence data‑center builders, underscoring the continued momentum of the AI infrastructure boom. CEO Christophe Fouquet highlighted that customers are increasingly confident about medium‑term AI‑related demand, suggesting that the industry expects sustained growth in chip production for AI workloads.Weiterlesen