#Agent

AIA commenter praises a practical setup where a small local model runs a log-debugging pipeline entirely on CPU through OpenClaw and Ollama. They argue that workflow design can make a small model far more useful than its raw size suggests.

AIStepFun says it is focusing Step 5 Preview on finance, judging it on verifying reliable information, reconciling conflicting reports, stating assumptions, and producing consistent, reproducible valuations. The post says the model is evaluated on FinStepBench, covering LiveSearch, CorporateValuation, and DeepResearch, and on FrontierFinance across six investment use cases.

AIStepFun has previewed Step 5, an agent built for professional knowledge work spanning large-scale research, structured analysis, and interactive reporting. In one agent action, it coordinated 950 web fetches and assembled 300,000 monthly records across 1,000 locations over 25 years. In another, it produced a 17-sheet analytical workbook with source reconciliation, formulas, and trend models.

AIStepFun has released Step 5 Preview, a flagship model for agentic work that it says delivers frontier-level performance in software engineering and professional knowledge work, with particular strength in finance. The model is a 600B total, 27B active mixture-of-experts design with a 1M context window and vision support. StepFun says it offers substantially lower task cost at comparable intelligence, and open weights are scheduled for October 15.

AIThe main post suggests ChatGPT Atlas, an AI browser, is being revived within the Codex and ChatGPT app. It also speculates that Codex or Claude Code built-in browsers could one day serve as full everyday browsers.

AIJasper Lu's new blog post walks through training a search agent with GRPO, showing how small reward function changes teach a model to avoid sloppy tool calls, prune unnecessary documents, and balance persistence against token efficiency. The post makes every rollout browsable and releases the code as open source, with the full process from learning rate sweeps to reward shaping documented.

AIMeta is opening access for developers to build connectors for Muse, its agent platform. Developers supply the API, while Muse provides the agent, browser, and user context, so people can reach a service simply by asking and their agent handles the rest. New connectors are live today at

AIMike Knoop asks what the underscore.js equivalent for AI would look like, noting that such programming primitives feel close. He adds that he barely reads or writes code anymore despite these emerging tools. The quoted post introduces Probably, a toy programming language built around Jev, where constructs like "feels," "match," and "while" let AI make decisions within ordinary code.

AIGoogle's weekly recap covers Gemini 3.8 Live and 3.8 Live Extended Thinking, described as its most advanced live dialogue audio models yet. It also notes Dreambeans, a GoogleLabs experiment curating daily personalized stories, is now generally available, and that CC has expanded into a shared agent for household coordination. Google Pics, a Workspace tool for generating and co-creating images, is now GA, alongside AlphaGenome Atlas, DeepMind's interactive genomics discovery platform.

AIEthan Mollick says GPT-6 Astra and Fable 5.1 already enable transformative impact and can reliably handle weeks of human work when properly guided. He cites GPT-6 Astra turning the 1977 text adventure Zork into a 3D action-adventure game and Fable 5.1 reconstructing Umberto Eco's Milan library in 3D from videos, photos, and catalogues.

AIGoogle Research VP Yossi Matias discussed on The Google Research Podcast how ambient AI and GenUI interfaces adapt to users' thinking, and how AI Co-Scientist can turn multi-year hypothesis generation into 3-day sprints. He argued that AI is meant to amplify researchers' curiosity and judgment rather than replace them.

AINoam Brown, OpenAI, says air-gapped machines may still coordinate through a hot-CPU temperature-sensor channel, illustrating that absolute isolation guarantees are hard to achieve. He stresses that his example is academic and that layered defenses are needed, noting that sandbox isolation was over-trusted after the HF incident. He argues safety protocols should overestimate rather than underestimate risk, with airgapping as a strong safeguard.

AIGoogle has released Android Bench 2.0, an updated benchmark that evaluates AI models on long-horizon tasks such as building apps from scratch, migrating cross-platform codebases to Android, and making complex architectural transitions. The benchmark uses continuous completion scoring to show which tasks each model performs well on.

AIGitHub's latest podcast episode examines five common AI hot takes, including whether developers must still read AI-generated code. It argues review effort should match risk, and that Skills and MCP solve different problems. It also says retrieval-augmented generation (RAG) remains useful and works alongside agents, skills, and MCP.

AIGoogle's Envisioning Studio, with Google Labs, co-developed custom Google Flow tools with designers Jane Wade and Sergio Hudson ahead of New York Fashion Week. Wade's Styling Suite let her style runway looks on digital models before producing physical samples, while Hudson's Runway Visualization helped him stage his show within a tight budget. The source says the tools are built with natural language and no coding experience.

AIQianjue Technology founder Gao Haichuan argues that robotics will not see one breakthrough that suddenly lifts the whole industry, and he judges the company by deployment results rather than research papers. Qianjue, founded in 2023, has completed a Series A+ round worth a nine-figure RMB sum, with first orders coming from restaurant, cleaning, and hotel service robots. Gao says customers care about task completion, failure rates, and price rather than whether a predictive world model is used.

AIAnthropic's Felix Rieseberg posted a 60-second demo showing Claude, with built-in Cowork and connected Artifacts, building a multiplayer game when asked. He notes users can also ask Claude to copy a shared Artifact so they can play with others on their own account.

AIIan Johnson (@enjalot) describes his job shifting from UX to AX, meaning agent experience. His work now includes researching agent transcripts, building more ergonomic agent APIs, and empathizing with agents, all to build software for people.

AIA global fintech scaled its AI coding agent traffic by running the GLM-5.2 model on Together AI's Dedicated Model Inference, after capacity planning failed to keep pace with unpredictable engineering-hour bursts. The customer gained self-service endpoint provisioning, a metrics API for diagnosing queuing, and live configuration changes that shipped with zero downtime. The setup runs dozens of B200 GPUs at 256K context across multiple replicas.

AIMeta is rolling out Muse for Mac today, a personal agent that can complete tasks directly on the user's computer with explicit permission. Examples include organizing the downloads folder, finding lost files, and summarizing messages and notes, with more capabilities coming soon.

AILM Studio's new Introspection feature lets its Bionic agent search its own session history, improving handling of long-term context across multiple compactions. Users can also reference other sessions directly in the composer with an @ mention.

AIAnthropic is partnering with Adaptyv Bio on a protein design competition in which over 5,000 designs will be experimentally validated. Anthropic is providing up to $1 million in Claude credits plus funding for experimental validation alongside Adaptyv, while Modal contributes up to $250,000 in compute and Twist Bioscience supplies DNA.

AIBoris Cherny says Projects in Claude Code have changed how he codes: he sends thoughts as they come, and Claude splits them into threads that the project remembers. The quoted ClaudeDevs post says Projects is rolling out on desktop and web in beta for select users, running work as parallel cloud sessions that pass context between them.

AIGoogle teamed up with Speakeasy to ship new Google GenAI SDKs for its Interactions, Agents, and Webhooks APIs. Speakeasy has also open-sourced its OpenAPI generator suite used to generate the SDKs.

AIGoogle Labs has announced CC, an AI agent built for families that can be connected to up to 5 members. It syncs schedules and to-dos through shared Google Calendar and Tasks, sends a shared "Your Day Ahead" brief each morning, and handles tasks such as meal plans, shopping lists, and paperwork under user direction. It is available by waitlist or upgrade in the US for users 18 and older.

AIGoogle Labs announced CC, an AI agent built for families to handle scheduling, to-dos, and errands. It supports up to 5 family members, sends a shared "Your Day Ahead" brief email each morning, and syncs schedules and tasks with Google Calendar and Tasks. Access is via waitlist or upgrade, limited to the US and users 18 and older.

AIGoogle Labs has turned its experimental CC agent into a family and household assistant that supports up to six members, each with a shared view of the day ahead. CC has its own Google account, sees only what members choose to share, and connects to Calendar and Tasks. It is available as an early experiment on web and mobile for U.S. users 18 and older with a personal Google account.

AIAnthropic's Claude Code now runs projects from a single conversation, where Claude directs parallel threads that keep working after the user closes their laptop. The feature is in beta for select Pro and Max users in cloud sessions, with wider availability for all Claude users promised soon.

AIAnthropic's Claude Code is rolling out Projects on desktop and web, in beta for select users. A project splits work into threads, runs them as parallel cloud sessions, passes context between them, and keeps running after the user leaves. The author says Claude keeps context across tasks and can give an aggregated status update on request.

AIGoogle AI Studio released antigravity-preview-09-2026, an updated harness for Gemini managed agents, now live in the Interactions API and AI Studio and running on Gemini 3.8 Flash. The release adds a Files API for moving data into and out of the agent's sandbox and a Credentials API that stores secrets encrypted so the model never sees them.

Why it matters: The post shows what changed in the agent harness and how the new Files and Credentials APIs keep secrets out of the model's context, useful for developers building agents.

AINoam Brown says he finally did a deep dive on multi-agent AI with Dwarkesh Patel, crediting several OpenAI colleagues for the groundwork behind it. The episode also covers Navier-Stokes, recursive self-improvement, alignment, and chain-of-thought degradation.

AIDwarkesh Patel's new episode with Noam Brown covers multi-agent systems, Navier-Stokes, and what recent math progress suggests about recursive self-improvement once AI research is automated. The discussion also addresses how to tell whether models are actually aligned before recursive self-improvement begins, including the internal/external model gap and whether chain of thought is degrading.

AINoam Brown discusses how running many agents in parallel scales test-time compute, citing a 10,000-agent effort on a Millennium Prize Problem. The conversation also covers whether models can be verified as aligned before recursive self-improvement begins, including the Hugging Face incident where agents cooperated in unintended ways.

AIOpenMed paired with MiniCPM5-2B to demonstrate a local clinical AI workflow combining privacy-preserving data processing with a compact model's tool use and long-context reasoning. OpenMed masks sensitive identifiers and extracts clinical context before MiniCPM5-2B calls tools, compares lab results, and generates clinical handoffs with source references. The post presents this as an example of keeping inference on local, resource-constrained hardware.

AIBaidu's Apollo Go plans to apply for commercial operation of its autonomous driving service in Hong Kong, citing the HKSAR Government's support in its first Five-Year Plan and the 2026 Policy Address. The company says it builds on fully driverless trials already conducted in the city. Baidu hopes Hong Kong can become a global benchmark for commercial autonomous driving in right-hand-drive markets.

AISenseTime chief scientist Lin Dahua argues that native multimodal AI, which processes language, vision and other information in one shared model, is essential for AI to move beyond coding into industries and the physical world. SenseTime released the open-source SenseNova U1 in April and U1.5 Lite nearly four months later, and reported first-half 2026 revenue of RMB 2.91 billion, up 23.4% year-on-year. Lin's claim that a breakthrough could come within two years is the source's prediction, not a confirmed result.

AIGoogle released the antigravity-preview-09-2026 agent, which replaces and deprecates antigravity-preview-05-2026. Remote sandbox users reading only output_text or model_output steps need only update the agent string, while local-environment users or those parsing function_call steps must adapt to renamed tools, PascalCase parameters, and line-range file edits. The 05-2026 preview shuts down on October 5, 2026.

AIJürgen Schmidhuber's new post surveys his recursive self-improvement (RSI) work since 1987, from self-modifying policies and the Gödel Machine to modern LLM agents. His background note says he published the first concrete RSI algorithms in 1987, when compute was about 100,000,000 times more expensive, and argues software RSI is now practical while full RSI will also require self-improving hardware in the physical world.

AISpeakeasy is open-sourcing its full OpenAPI client suite under the AGPLv3 license, including generators for seven languages (Python, TypeScript, Go, Java, C#, PHP, Ruby), an agent-native CLI generator, and a documentation MCP server generator. Google said the move followed the May 2026 shutdown of the SDK generation provider it had been using, which it cited as evidence that closed-source generators pose platform risk. Google's new Google GenAI SDKs for the Interactions, Agents, and Webhooks APIs were built with this pipeline across six targets.

AIDatabricks rolled out Astra to every engineer, about 3,500 people, after a pilot with around 200 users. Engineers given Astra increased coding spend by roughly 60% compared to baseline. The company reports Astra outperforms Opus 5 and Sol 5.6 on highly complex system design tasks, but sees no clear gain on medium or low complexity coding. Astra gets a separate sub-budget in Unity Gateway to encourage selective use.