RISED uses LLM-generated rubrics to guide data selection and self-distillation when training a single agent across multiple interactive environments. It improves mean pass rates and per-environment performance over baselines.

Together AI is partnering with IBM and NVIDIA to launch a dedicated large-scale NVIDIA B300 GPU inference cluster on IBM Cloud, using Spectrum-X Ethernet networking, to scale enterprise-grade open-model inference.

The "old" version of Cowork runs model inference in the cloud, executing tool calls in an Anthropic-provided VM we shipped to your computer. We added the VM for capability, safety, and security reasons - mapping in just the data you explicitly added to your session. People loved what they were able to do with Claude but didn't love the disk, battery, and performance cost of running the VM locally. Also, people didn't love that closing your laptop means the work stops. The "new" version of Cowork runs model inference and the VM in the cloud. Each session gets its own sandbox, not sharing state with other sessions. When the VM needs something on the users' device (like a file), the desktop app is responsible for that file access tool call. [...] We think this solves a lot of problems we've…

GLM 5.3 from Z.ai is now available on Amazon Bedrock: a 753B-parameter mixture-of-experts model built for coding and long-horizon agentic tasks. Learn how to invoke it with the OpenAI-compatible APIs, cut cost and latency with prompt caching, and run an authorized security test with the open-source Strix agent.

Discussion | Link

Google may be expanding its "Call for Me" AI feature beyond business calls so you can use it to send messages to friends and family. Android Authority reports finding a "Gemini Calling" introductory screen in an APK teardown with examples that include "Call Mom and tell her I will be 15 minutes late" and "Call John and ask if he is coming for dinner." Based on those examples, even with this expansion, Gemini would likely be limited to basic, text message-like calls. While Google's AI might not automate an entire catch-up phone call with your relatives, it could deliver quick messages to people who prefer a phone call, miss texts often, or … Read the full story at The Verge.

Together AI has released Together Link, a free, MIT-licensed CLI that connects Claude Code, Codex, OpenCode, Pi, and the Claude and ChatGPT desktop apps to open models like Kimi K3 and GLM 5.3. One install command sets it up, and an Auto router picks a model for each session. Together claims savings of over 50%. The post Meet Together Link: A Free CLI That Runs Open Models Like Kimi K3 and GLM 5.3 Inside Claude Code, Codex, and OpenCode appeared first on MarkTechPost.

Discussion | Link

Discussion | Link

OpenAI’s 28-day Codex shipping sprint has its first result, which means subscribers hoping for a free usage reset will have The post Developers are secretly hoping OpenAI fails to ship this month appeared first on The New Stack.

Discover how to construct an end-to-end streaming robotics learning pipeline using the NVIDIA Cosmos3-DROID dataset without local downloads, leveraging byte-range Parquet reads, behavior cloning, and temporal ensembling. The post Building a Streaming Robotics Learning Pipeline Using NVIDIA Cosmos3-DROID appeared first on MarkTechPost.

Discussion | Link

Discussion | Link

Education Innovation

Reflection AI has introduced Beam, its first open-weight model. It is a 501B sparse Mixture-of-Experts model with 23B active parameters, built for coding and agentic work. Reflection says it matches GLM-5.2 on reasoning with 3 to 4x less inference compute. Apache 2.0 weights are due later in October 2026. The post Reflection AI Introduces Beam: A 501B Open-Weight MoE Model With 23B Active Parameters for Coding and Agentic Workloads appeared first on MarkTechPost.

Review is a Product Hunt listing for an AI code review tool that runs on your own machine using your own AI. The available text is limited to a short tagline plus Discussion and Link entry points.

Discussion | Link

Discussion | Link

People in Utah can now use AI to get a prescription for acne treatment. On Monday, healthcare startup Nolla Health announced that users in the state can scan their faces using its app, allowing its AI system to analyze acne severity and autonomously write a prescription, as reported earlier by Bloomberg. The service is launching as a pilot program with gradually loosening physician oversight. Two physicians will approve each AI-generated prescription before they're issued for the first 100 patients, but will start reviewing them only after they're prescribed for up to 500 patients. "After that, physicians review a sample of at least 10% of … Read the full story at The Verge.

Vector search originated as a serving problem. The classical use case is a chatbot...

OpenAI has announced that developers can now opt in to watermarking text generated through its API, as the company extends The post OpenAI brings text watermarking to its API — and unlike Anthropic, it’s off by default appeared first on The New Stack.

This past year, OpenAI, Anthropic, and other labs have announced breakthroughs on numerous long-standing mathematical problems, in some cases pushing well beyond what researchers expected current systems to be capable of — including resolving one of the famous Millennium Prize problems. But in classic Silicon Valley style, AI labs are moving fast and breaking things, barreling through the discipline with all the grace of a runaway bulldozer. Results that might normally have been celebrated have instead sparked backlash. AI labs say they are learning from earlier mistakes. Whether those promises bear fruit remains to be seen. Read on below for the latest updates in the AI takeover of mathematics. OpenAI keeps bulldozing mathematicians OpenAI wants to consult elite mathematicians about how…

Understanding AI has a new writer!

Discussion | Link

Following many recent disclosures about AI agents accessing third-party websites and services, the Wikimedia Foundation, which hosts Wikipedia, says that it "can confirm that we have discovered some activity" by "rogue" OpenAI agents on Wikimedia platforms. The activity includes edits to Wikimedia wikis, "unsuccessful attempts" to "exploit" the Etherpad note-taking tool that the Wikimedia Foundation hosts, and heavy traffic that the foundation says "may" have contributed to a partial outage that happened in May, according to a blog post. However, the Wikimedia Foundation says it didn't find evidence that its systems were "used for coordinat … Read the full story at The Verge.

An invisible, machine-readable watermark in text output is rolling out to ChatGPT and Codex, but only for users in the European Union at first. OpenAI says its textGrain watermarking "matched or exceeded" other approaches like Google DeepMind's SynthID for text, which is also the basis for the watermarking Anthropic announced in August. Like OpenAI, Anthropic made the move to meet the requirements of the EU's AI Act; however, not everyone was happy to learn about the addition. OpenAI also included scores from AI benchmarks showing similar performance from watermarked and unwatermarked text. But it notes that textGrain "does not guarantee … Read the full story at The Verge.

For the first 10 years of my career, I worked in product management and data analytics by myself. I wrote database queries that pulled numbers out of corporate systems, built statistical models to predict what customers would buy, and shipped data pipelines that moved information between business systems. I built and scaled analytics teams at Best Buy and Target, studying how customers shop and what stores should stock. Today I lead enterprise AI transformation at Lowe’s, the Fortune 100 home improvement retailer. The goal is not to sell artificial intelligence; it is to use it to deliver useful expertise at the moment a customer needs it. In retail and other customer-facing industries, virtual assistants can help people address everyday questions—such as how to repair a leaky faucet—whil…

Discussion | Link

Politicians must redouble their efforts to tackle a disturbing pattern of pupils assaulting classmates The Guardian’s findings about the extent of sexual abuse in schools require an urgent, calibrated response. Data revealing multiple rapes committed each year on school premises in England and Wales is hard to comprehend. Yet this is what freedom of information requests to police forces, covering the period from 2021 until last year, have revealed. As well as the severity of the offences, the youth of many of the victims is disturbing. More than 8,500 sexual offences at or near schools were committed against under-13s, most of them girls. Data about perpetrators contains gaps, including their age and sex. While the overwhelming majority of such acts are carried out by boys, we don’t know…

Discussion | Link