IBM Research Releases CUGA: Open-Source Toolkit for Building AI Agent Apps
IBM has launched CUGA, an open-source framework for creating AI agent applications. It includes two dozen working examples to help developers get started quickly.
#ai
AI
2101 stories tagged AI · page 48 of 88
GPT-5 Solves 3-Year Immunology Mystery, Advancing Cancer Research
GPT-5 Pro helped immunologist Derya Unutmaz solve a three-year-old puzzle about how T cells behave, specifically regarding an unusual subset of T cells. This breakthrough could accelerate research into cancer and autoimmune diseases.
FirstPass: Grounding AI Scientific Judgment in Multi-Round Editorial Outcomes
Researchers created FirstPass, a dataset and AI model trained on real multi-round peer-review dialogues from Nature Communications. This could make AI evaluations of scientific papers more accurate by learning from actual editorial decision-making, not just stylistic mimicry.
Fika Jobs raises $4M to build a video-first hiring platform where AI agents interview candidates
Stockholm-based Fika Jobs has raised $4 million to build a video-first hiring platform that combines AI interview agents with short-form video profiles, creating a cross between LinkedIn and TikTok for job applications.
Darwin Mobile Agent: A Roadmap for Self-Evolving AI
A new paper proposes a roadmap for AI that learns through interaction with complex environments, mimicking natural evolution by systematically removing human priors. The authors have released an open-source infrastructure called Darwin Mobile Agent to test this idea using mobile graphical user interfaces as a practical proxy for an open-ended world.
AlphaMemo: AI Agents That Learn from Their Financial Trading Mistakes
Researchers have developed AlphaMemo, an AI system that helps financial trading agents learn from their past decisions. This could make AI-driven trading more reliable and less prone to repeating errors.
AI Models Still Make Irrational Decisions Even When Aligned
Researchers found that AI models can still make irrational decisions even when trained to align with human values. This 'rational value risk' means models may not always choose the best possible action, even if they understand what's valuable.
Voice AI Chess Companion Helps Elderly Players Stay Sharp in 3 Languages
A developer created an AI voice assistant to help older adults play chess. It supports English, Spanish, and French, making the game more accessible.
SpaceX and AI Startup Reflection Ink Up to $6.3B Computing Power Deal
SpaceX has partnered with AI startup Reflection to build a massive data center. This deal could accelerate AI development and space exploration.
Samsung Rolls Out ChatGPT and Codex to Global Workforce
Samsung Electronics has deployed ChatGPT Enterprise and Codex to its employees worldwide. This is one of the largest enterprise AI rollouts for OpenAI's tools.
PP-OCRv6: New Open-Source AI Can Read 50 Languages in Tiny Devices
PaddlePaddle released PP-OCRv6, an AI that can read text from images in 50 languages. It's up to 700% more efficient than its predecessor, working even on smartphones. This makes it easier to digitize books, signs, and documents in any language.
Nvidia's New Cooling System Saves Water, but Misses AI's Bigger Problem
Nvidia introduced a cooling system that reduces water use in data centers. However, this doesn't address the larger issue of water consumption by fossil fuel power plants that supply AI infrastructure.
In the Weights is your new AI-centric vanity search
In the Weights is a new service that lets you see how AI models perceive you. It's like a vanity search, but for AI-generated insights.
Google DeepMind and A24 Partner to Build AI Filmmaking Tools
Google DeepMind is investing $75 million to collaborate with A24 on AI tools for filmmaking. This partnership could bring AI assistance to areas like scriptwriting and editing.
GitPT: AI-Generated Git Commit Messages Using Apple's On-Device Models
GitPT is a new open-source tool that automatically generates Git commit messages using Apple's on-device AI models. This means you can get smart, professional suggestions without sending your code to the cloud.
DeepMind Publishes AI Control Roadmap to Safeguard Future Autonomous Agents
DeepMind has published a detailed technical roadmap for safely controlling advanced AI agents. The plan addresses security failures, robust monitoring, and human oversight to ensure AI remains beneficial as it evolves.
AI-Powered Music Generation: Create Songs in Seconds
A new AI tool lets anyone generate short music clips from text prompts, sparking debate about the future of music creation and copyright.
AI Chipmaker Groq Confirms $650M Raise, Re-Staffs After Nvidia's Multi-Billion Dollar Deal
Groq confirmed a $650 million funding round after Nvidia's reported $20B deal that effectively acqui-hired a chunk of its leadership and engineers, rather than a full acquisition. Groq is now expanding its neocloud business and hiring new executives to refill its ranks.
You're Probably Overpaying for AI Tasks That Don't Need GPT
Many people use expensive AI models like GPT for simple tasks that cheaper or free alternatives can handle just as well. Switching to the right tool can save you money and improve performance.
Nobel Laureate John Jumper Is Leaving DeepMind for Rival Anthropic
John Jumper, a Nobel Prize-winning scientist, is moving from Google DeepMind to rival AI company Anthropic. This is part of a larger exodus of talent from DeepMind, as several other key researchers have also left. This shift could impact the future of AI development and competition in the field.
Snap Spins Off AI Video Team into New Company, Dotmo, Citing Costs
Snapchat's parent company is spinning off its AI video team into a new venture called Dotmo. The move highlights the high costs of AI development and Snap's strategy to focus on its core business.
Signal’s Meredith Whittaker wants you to remember that AI chatbots ‘are not your friends’
Meredith Whittaker, Signal’s president, warns that AI chatbots are not conscious beings and treating them as friends can be misleading. She emphasizes the importance of understanding their limitations and potential risks.
Jane Street Quietly Builds One of Wall Street's Most Advanced AI Systems
Jane Street, a powerful but low-profile trading firm, has developed cutting-edge AI tools that are transforming Wall Street. Their systems handle complex trading decisions faster and more accurately than human traders.
I'd Rather Risk Cancer Than See AI Move This Fast
A provocative new opinion piece argues that the rapid pace of AI development poses greater existential risks than slowing down progress on medical breakthroughs like cancer cures. The author frames the debate as a stark trade-off between technological acceleration and societal safety.