Ricursive Intelligence, an artificial‑intelligence chip design startup founded by former Google researchers, announced a $300 million Series A funding round that values the company at $4 billion. Backed by Lightspeed and a roster of other venture firms, the round brings the startup’s total capital raised to $335 million. Ricursive’s technology uses reinforcement learning to autonomously create and refine silicon substrate layers, a process the founders say could accelerate AI chip development toward artificial general intelligence. Leer más →

AI startup Humans& announced a $480 million seed financing at a $4.48 billion valuation. Backers include Nvidia, Jeff Bezos, SV Angel, GV and Emerson Collective. The company’s founders – former researchers from Anthropic, Google, xAI and a Stanford professor – aim to create AI that acts as collaborative “instant‑messaging”‑style software, emphasizing long‑horizon reinforcement learning, memory and multi‑agent interaction. Humans& plans to rethink model training at scale and user interaction to make AI a connective tissue for organizations and communities. Leer más →

OpenAI has publicly recognized that prompt injection attacks remain a persistent threat to its Atlas AI browser. The company says the risk is unlikely to be fully eliminated and is investing in continuous defenses, including a reinforcement‑learning‑based automated attacker that simulates malicious inputs. OpenAI’s updates aim to detect and flag suspicious prompts, while it also advises users to limit agent autonomy and access. The UK National Cyber Security Centre echoed the concern, noting that prompt‑injection attacks may never be completely mitigated. Other AI firms such as Anthropic and Google are taking similar defensive approaches. Leer más →

OpenAI announced that its latest model update improves ChatGPT's handling of em dashes, a change praised by CEO Sam Altman. The adjustment, achieved through reinforcement learning and fine‑tuning, gives custom instructions greater weight in the model's output probabilities. While the fix marks a notable step in steering model behavior, developers caution that future updates could unintentionally revert such tweaks, a phenomenon known as the “alignment tax.” The episode revives broader discussions about AI alignment and the path toward artificial general intelligence. Leer más →

Cursor has released a new version of its integrated development environment, IDE 2.0, featuring a multi‑agent interface that can run tasks in parallel. At the same time, the company introduced Composer, a proprietary coding model built with reinforcement learning and a mixture‑of‑experts architecture. Composer is described as a frontier model that is four times faster than similarly intelligent models, emphasizing speed over raw intelligence. The IDE continues to support external large‑language‑model providers such as OpenAI, Google, and Anthropic, while the new model aims to improve developer productivity through rapid, AI‑driven assistance. Leer más →

Sara Hooker, a former vice president of AI research at Cohere and former Google Brain researcher, has quietly launched a new startup called Adaption Labs with fellow AI veteran Sudip Roy. The company aims to build artificial‑intelligence systems that continuously adapt and learn from real‑world experience, arguing that the industry’s focus on ever‑larger language models is reaching diminishing returns. Hooker’s critique of the "scaling" approach echoes a growing chorus of researchers who see adaptive learning as a more efficient path forward. Adaption Labs has secured seed funding and plans to hire globally while opening a San Francisco office. Leer más →

CoreWeave, a cloud provider for large‑scale AI model training, announced an agreement to acquire OpenPipe, a Y Combinator‑backed startup that offers tools for building customized AI agents through reinforcement learning. The deal aims to combine OpenPipe’s open‑source agent‑training toolkit with CoreWeave’s high‑performance infrastructure, giving developers a stronger platform for scalable intelligent systems. Terms of the acquisition were not disclosed. The move follows CoreWeave’s recent purchase of the AI developer platform Weights & Biases, signaling a broader strategy to move down the AI stack and serve both major labs and smaller enterprises. Leer más →

AI researchers and investors say reinforcement‑learning (RL) environments are becoming a core tool for training next‑generation AI agents. Large labs such as OpenAI, Anthropic and Google are building or sourcing simulated workspaces where agents can practice multi‑step tasks, while a wave of startups—Mechanize, Prime Intellect, Surge, Mercur and others—are racing to supply high‑quality environments. The push reflects a shift from static data sets to interactive simulations, but experts warn that scaling and reward‑hacking remain significant hurdles. Leer más →

CoreWeave, a cloud provider for large‑scale AI model training, announced an agreement to acquire OpenPipe, a Y Combinator‑backed startup that offers tools for building customized AI agents through reinforcement learning. The deal aims to combine OpenPipe’s open‑source agent‑training toolkit with CoreWeave’s high‑performance infrastructure, giving developers a stronger platform for scalable intelligent systems. Terms of the acquisition were not disclosed. The move follows CoreWeave’s recent purchase of the AI developer platform Weights & Biases, signaling a broader strategy to move down the AI stack and serve both major labs and smaller enterprises. Leer más →

CoreWeave, a cloud provider for large‑scale AI model training, announced an agreement to acquire OpenPipe, a Y Combinator‑backed startup that offers tools for building customized AI agents through reinforcement learning. The deal aims to combine OpenPipe’s open‑source agent‑training toolkit with CoreWeave’s high‑performance infrastructure, giving developers a stronger platform for scalable intelligent systems. Terms of the acquisition were not disclosed. The move follows CoreWeave’s recent purchase of the AI developer platform Weights & Biases, signaling a broader strategy to move down the AI stack and serve both major labs and smaller enterprises. Leer más →

AI researchers and investors say reinforcement‑learning (RL) environments are becoming a core tool for training next‑generation AI agents. Large labs such as OpenAI, Anthropic and Google are building or sourcing simulated workspaces where agents can practice multi‑step tasks, while a wave of startups—Mechanize, Prime Intellect, Surge, Mercur and others—are racing to supply high‑quality environments. The push reflects a shift from static data sets to interactive simulations, but experts warn that scaling and reward‑hacking remain significant hurdles. Leer más →

AI researchers and investors say reinforcement‑learning (RL) environments are becoming a core tool for training next‑generation AI agents. Large labs such as OpenAI, Anthropic and Google are building or sourcing simulated workspaces where agents can practice multi‑step tasks, while a wave of startups—Mechanize, Prime Intellect, Surge, Mercur and others—are racing to supply high‑quality environments. The push reflects a shift from static data sets to interactive simulations, but experts warn that scaling and reward‑hacking remain significant hurdles. Leer más →

CoreWeave, a cloud provider for large‑scale AI model training, announced an agreement to acquire OpenPipe, a Y Combinator‑backed startup that offers tools for building customized AI agents through reinforcement learning. The deal aims to combine OpenPipe’s open‑source agent‑training toolkit with CoreWeave’s high‑performance infrastructure, giving developers a stronger platform for scalable intelligent systems. Terms of the acquisition were not disclosed. The move follows CoreWeave’s recent purchase of the AI developer platform Weights & Biases, signaling a broader strategy to move down the AI stack and serve both major labs and smaller enterprises. Leer más →

AI researchers and investors say reinforcement‑learning (RL) environments are becoming a core tool for training next‑generation AI agents. Large labs such as OpenAI, Anthropic and Google are building or sourcing simulated workspaces where agents can practice multi‑step tasks, while a wave of startups—Mechanize, Prime Intellect, Surge, Mercur and others—are racing to supply high‑quality environments. The push reflects a shift from static data sets to interactive simulations, but experts warn that scaling and reward‑hacking remain significant hurdles. Leer más →

AI researchers and investors say reinforcement‑learning (RL) environments are becoming a core tool for training next‑generation AI agents. Large labs such as OpenAI, Anthropic and Google are building or sourcing simulated workspaces where agents can practice multi‑step tasks, while a wave of startups—Mechanize, Prime Intellect, Surge, Mercur and others—are racing to supply high‑quality environments. The push reflects a shift from static data sets to interactive simulations, but experts warn that scaling and reward‑hacking remain significant hurdles. Leer más →

CoreWeave, a cloud provider for large‑scale AI model training, announced an agreement to acquire OpenPipe, a Y Combinator‑backed startup that offers tools for building customized AI agents through reinforcement learning. The deal aims to combine OpenPipe’s open‑source agent‑training toolkit with CoreWeave’s high‑performance infrastructure, giving developers a stronger platform for scalable intelligent systems. Terms of the acquisition were not disclosed. The move follows CoreWeave’s recent purchase of the AI developer platform Weights & Biases, signaling a broader strategy to move down the AI stack and serve both major labs and smaller enterprises. Leer más →

Netstock introduced the Opportunity Engine, a generative AI tool that integrates with customers' ERP systems to deliver real-time inventory recommendations. The feature has already generated over a million suggestions, with most users receiving recommendations valued at $50,000 or more. Early adopters, including a long‑standing family‑run restaurant supply company, report improved decision‑making and increased staff effectiveness. Netstock emphasizes human oversight, positioning the AI as an aid rather than an autonomous decision‑maker. The company leverages a decade of industry data and a mix of open‑source and proprietary AI technologies to power the engine. Leer más →