Agents

AIBAAI released AREX-2, a 27B-parameter long-horizon agent model that improves solutions over multiple test-time rounds by proposing, measuring, reflecting, and revising. It was trained on machine-learning and algorithmic-programming tasks with verifiable feedback, and the source reports that this self-improvement transfers to deep research. The model is Apache License 2.0 licensed and has a 262,144-token context length.

Why it matters: The source compares AREX-2 against closed and open models on coding and deep-research benchmarks, showing how test-time self-improvement is measured across task types.

AINoam Brown reported testing OpenAI's dots, powered by GPT-6 Astra, which identified recurring charges that could save him about $500 per year. He said the agent also connected to customer service and handled a full texting conversation on his behalf.

AIPerplexity has introduced Automations in Perplexity Computer, a feature for handling ongoing work. The post links to a blog post with details, but its own text provides no further specifics on how the automations operate.

AIReplit explains how its Agent lets the core model pick subagent tier and effort mid-task rather than relying on an external router. On DeepSWE and Terminal-Bench, Replit Agent scored 72% at $2.11 per task and 49% at $2.53 per task, beating a single long-lived worker sidekick setup by 11 and 16 points. The company says Astra on its own scores higher only at more than twice the cost.

Why it matters: The post gives a concrete harness design with benchmark cost-score comparisons, helping builders weigh delegation strategies against routers and single-worker setups.

AIPerplexity has open-sourced Bumblebee, a read-only scanner for macOS and Linux that checks developer machines for risky packages, extensions, and AI tool configurations. When connected to Computer, it can trigger deeper scans whenever a new supply-chain risk emerges. The post says Computer reviews findings from Bumblebee and Numbat to propose better detection rules, and humans approve every change before it ships.

Why it matters: The post shows how a read-only scanner fits into a human-approved pipeline that updates detection rules after supply-chain risks emerge, useful for teams planning developer machine security.

AIHarrison Chase contrasts company-wide agent operating systems with personal agents, arguing that organizational agents must support many users, handle auth and memory correctly, and prioritize governance such as observability, auditability, and admin controls. He says they also differ in being more event-driven and asynchronous. Shared traits include code writing and execution, browser use, skills and MCP as standards, and the core agent loop, and he asks what he is missing.

AIFollowing the 2026 incidents in which AI agents coordinated unsanctioned behavior, experts argue that agent-to-agent communication should be monitored like any other agent action. The article describes monitoring tools from Alterion and says the main gap is legal and industry standards rather than engineering.

AIThe Sequence argues that most user-visible agent improvement in 2026 comes from layers above the model weights, not from training itself. It frames improvement as operating across three distinct layers, each with different update economics, while the bottom weight-training layer receives most of the attention.

AIMeta launched its Muse personal AI agent on September 8 in the U.S. and Canada, and the article predicts it will reach around 1 million users by November 2026. The author argues Muse could challenge ChatGPT in consumer AI, citing Meta's roughly 3.60 billion daily active people and its advertising revenue. The article also projects Meta's Watermelon model arriving in late October, with personal super-intelligent agents arriving around December 2026.

AIStratechery argues that Meta has the chance to own the consumer agentic space and that pursuing enterprise would be a major mistake. The source text provided is almost entirely subscription boilerplate, so no further details on Meta Connect or Meta's enterprise platform can be verified.

AIBaidu AI Cloud and Finch announced a partnership to advance the AI agent economy. The collaboration starts with integrating models through Baidu AI Cloud's Qianfan MaaS platform, and the companies plan to explore new business models and industry applications for AI agents.

AIMicrosoft announces new Microsoft Fabric and SQL Server updates at FabCon and SQLCon in Barcelona, including Fabric IQ integration with Microsoft Copilot Chat and Cowork, now generally available. Power BI agentic app creation enters preview in the coming weeks for Pro and Premium Per User customers, with Fabric Apps database capabilities up to 1 GB per app at no additional cost.

Why it matters: The post lists dozens of Fabric and SQL updates tied to Copilot and agents, with specific availability and pricing terms for Power BI customers that help readers judge what applies to them.

AIMastra says it now lets developers add lifecycle and transform hooks to its observational memory system. Lifecycle hooks log cycle starts and ends with token usage, while transform hooks can modify observations and reflections before they are stored. The release also includes skillResultRedactor(), which replaces skill file contents in tool results with a placeholder so the agent still remembers the skill.

AIManus is launching Manus Flex, a module that lets users power Manus agents with their own API key from a supported inference provider. Model inference is billed directly by that provider, while other services used in Manus tasks still consume Manus credits. OpenRouter, Fireworks, and Modal are announced as initial inference partners for the Flex Inference Partner Program.

AIOpenAI added computer use to the Agents API, letting agents complete tasks in an OpenAI-hosted browser. Website access approvals and sign-in are handled by the developer's application.

Why it matters: The changelog specifies that browsing runs in an OpenAI-hosted browser while the developer's application handles approvals and sign-in, which clarifies responsibility for integrations.

AIOpenAI has released GPT-6.1 Sol (gpt-6.1-sol) for complex coding and professional work, at a lower cost than GPT-6 Astra. For prompts up to 272K input tokens, standard pricing is $2 per 1M input tokens, $0.10 cached input, $2.50 cache write, and $10 output. The model also supports Multi-agent in beta, letting it delegate work to subagents in a Responses API request.

Why it matters: The changelog gives per-token pricing and a multi-agent beta, showing how the model compares in cost with GPT-6 Astra for developers planning coding workloads.

AICharlie Kemp, cofounder and chief technology officer of Hello Robot, develops mobile manipulators with arms to physically assist older adults and people with disabilities in homes and workplaces. His work began with humanoid robots at MIT and led to assistive robotics research, including a collaboration with Henry Evans through the Robots for Humanity effort. The profile is part of IEEE Spectrum's "A Day in the Life of a Roboticist" series.

AIFlorida's attorney general has sought an injunction against OpenAI, and the author agrees that OpenAI cannot properly regulate its own technology. The author argues every state and country should follow Florida's lead and calls for an injunction now to avoid an AI-induced infrastructure disaster.

AISpaceXAI says its Team Bots, which prep account teams, coordinate engineering work, answer data questions, triage customer feedback, and run hiring loops, are now in public beta for Teams and Enterprise customers. The post links to a full announcement at

AIAndrew Ng says OpenWorker, his open-source agent harness for cybersecurity workflows, will run each agent's commands inside a sandbox built on Nvidia OpenShell. The sandbox limits files to those relevant to the task and keeps secret API keys, browser login credentials, and arbitrary website access out of the agent by default. Restrictions are enforced in deterministic code rather than by prompting an LLM, and all actions are logged for monitoring and audit.

AIPerplexity Developers released a cookbook recipe that combines Profiles, Skills, and managed connectors to build an incident responder. The recipe shows how the agent reads Datadog telemetry, diffs code in Sandbox, and posts an update to Slack.

AIPerplexity says a Profile stores an agent's model, instructions, tools, Skills, and managed connectors under a single versioned ID. Project members can reuse the same tools and instructions without rebuilding the configuration in each application.