Next Token episode 5 discusses why Dots is hard to use and why personal agents may not succeed in China. The hosts also examine the impact of AI being able to replicate any software, along with opportunities from small hardware devices.
This is pretty sick
Tessl's blog post argues that teams should record agent mistakes as attributed, signed diary entries, then curate them into reusable context packs rather than adding unverified rules to files like AGENTS.md. The author describes a REST API case where an agent regenerated the OpenAPI spec and TypeScript client but missed the Go client, and the same lesson had to be re-taught in a fresh session.
AI DevCon New York, running November 2–4 at Industry City in Brooklyn, centers its program on software factories, the systems needed to make agentic development repeatable, trustworthy and scalable. The article argues that moving from one developer using an agent to an engineering organization requires layers covering context and skills, harnesses and tools, orchestration, verification and evaluation, and feedback.
The article explains how a software factory combines agents, tools, and engineering practices to take work from intent to verified output. It argues that teams should enforce domain rules, workflow gates, and tool permissions independently, version shared guidance and checks, and review lessons before reuse.
Zhang Lei, then a recent PhD graduate working mainly on image retrieval, beat a team led by face recognition expert Li Ziqing in a 2002 Microsoft Research Asia contest to build a face detection system. He cut the false-positive rate from roughly 50% to 1%, and the algorithm was integrated into Windows. Several members of his team later became prominent figures in China's technology industry.
Infor is building industry-specific AI agents on its Infor OS foundation and open architecture, according to CEO Kevin Samuelson. Infor says two in three businesses find off-the-shelf AI does not adequately address their industry's needs. The company pairs customers with forward-deployed engineers, and Samuelson says prototypes can now take one to three weeks.
Infor is developing industry-specific AI agents for industrial manufacturing, aerospace and defense, automotive, and food and beverage, built on its existing industry applications. Suresh Jayaraman, Infor's senior vice president of product management and development, said generic agents often fail to give deterministic answers and produce hallucinations. Infor's 2026.10 release adds guardrails through security and scopes, while agents still need human approval to move orders between customers.
Contact center success is shifting from call speed and deflection toward resolution, with experts arguing that AI agents should be measured by "conversation to completion." Speakers at theCUBE's coverage said AI can handle high-volume, low-stakes calls while humans take complex issues, and that context must carry across AI-to-human handoffs.
Midwest Wheel Companies is building toward AI agents that fix problems, with senior vice president Steve McEnany saying the company ties new capabilities to a single Infor system as a point of reference. The company's AI features include a product recommender in order entry and automation that scans emailed PDF invoices into the system. McEnany said a human should stay in the loop for anything touching cash or accounts, and that data and rules must be checked before agents act.
Wall Street is increasingly letting AI agents interpret market information, so a bad source or missing context can spread into real-world decisions before anyone detects the error. The source article does not identify specific models, products, or institutions.
Personal AI agents can sometimes overshare, overpromise and overpurchase, according to the article's summary. The source text is brief, and it does not name specific products, companies or incidents.
Tessl's talk at AI Native DevCon London argues that agent skills, which can be markdown files with instructions and bundled material, act as supply chain components that can shape agent behavior. The author says reading SKILL.md once is insufficient because risks can sit in supporting files, updates, and workspace trust settings. He identifies the danger as the combination of private context, untrusted content, and external communication, and cites research scanning roughly 4,000 public skills for issues including malware-like behavior.
Tessl argues AI code review is slow because generated code outpaces trust, and proposes executable specs that let agents check preview environments against product intent. Its spec reviewer splits work between a planner agent that extracts requirements and parallel verifier agents that test each one against the code and base branch.
The busywork around every demo nearly disappeared: Prep: 30 min → 5 Notes: 15 min → 2 CRM: 30 min → 7 That's about an hour back per call. Every rep can now take ~2 more calls a day and show up just as prepared.
Liquid AI is building personal AI that runs on devices such as phones, wearables, PCs, and cars, using its Liquid Context layer, which is optimized for Snapdragon processors, to sit between models, agents, and hardware. The company's agent harness uses its own models to decide which user context to retain and how to compress it within fixed compute limits. Liquid AI is also collaborating with Mercedes-Benz Group AG to bring on-device AI to its cars and plans observability and continuous improvement loops for self-improving agents.
Tessl's blog post argues that repository automation needs Continuous AI, a third pillar alongside CI and CD for scheduled, auditable AI workflows that improve repositories over time. The article describes GitHub Agentic Workflows, which harden agentic workflow specifications into GitHub Actions that can run coding agents such as Claude Code, Copilot CLI, Gemini CLI, or Codex-style agents. It emphasizes read-only agent steps, restricted outputs, and human review of pull requests.
Natura AI's Interface is a ring that lets users press and hold to speak requests to AI agents such as Claude Code, Codex, or Hermes, then release to send them. The post argues that screenless interfaces may define the next phase of agent use, since handing work to agents is currently slowed by pulling out a phone. Early-adopter pricing is $99, with shipping slated for January.
Satya Nadella outlines Microsoft's vision of Copilot as a new operating system for work spanning every model and task, backed by a governed "headless" business layer. Microsoft announced over 30 new Copilot skills across Dynamics 365 Sales, Service and Customer Insights, plus Microsoft Copilot Managed Runtime for hosting code inside a company's IT-governed environment. Nadella describes this as an "infinite SaaS factory" where users can describe needs and build customizations connected to existing systems of record.
The post promotes a curated collection of recursive self-improvement (RSI) papers covering the field's history and latest techniques. The author describes RSI as a hot, still-early AI topic and links to the collection at academy.dair.ai.
A Kyndryl survey of 2,000 senior IT decision-makers found only 10% are applying agentic AI as a modernization tool, and nearly half report being behind schedule with cost overruns. Researchers say agentic AI shows early promise for mapping hidden dependencies, generating code, and creating documentation, while Andy Thurai, a former IBM chief strategist, warns that AI-driven infrastructure sprawl could make compute costs unpredictable.
The author argues that agents working across a company fail because they lack the decisions and context recorded in threads, meetings, and DMs, not because the model is weak. The approach stores distilled claims with source evidence and time, never overwrites facts, labels missing information explicitly, and resolves permissions before the model runs. The report cites results on LongMemEval, including 99.8% top-ten evidence recall and $8.24 ingestion cost, and says an open-weight model can match frontier extraction quality.
Teamily AI now lets users add people and AI agents to the same group chat, so everyone works from shared context. The example shows a branding change handled by research, writing, and website-building agents, with a designer's feedback incorporated and the finished page shared in one continuous conversation. The platform's 2.0 release, which the post describes as opening to everyone, adds real-time human–agent collaboration and multi-model routing.
AI-native tools are removing the data bottleneck that long constrained CFOs, shifting the role toward designing the operating systems that turn data into decisions. Finance teams are adopting AI-native software for ERP, forecasting, procurement, and audit, and "finance engineers" are building custom automations and agents. OpenAI's CFO Sarah Friar describes finance moving toward a zero-day close and continuously updated forecasts.
Meta's Muse and OpenAI's Dots are always-on AI agents with animated mascots, pitched to consumers and businesses for tasks like restaurant reservations and inbox triage. Muse is free, while Dots is not, and OpenAI also offers "specialist" Dots for marketing, legal work, and accounting. The discussion centers on privacy and security concerns about giving agents access to credit card details and email.
I also sometimes think of why I'm working more, not less, since there are AI tools? From @samnewman: "AI was supposed to free us from drudgery. It isn't for most software developers because we're doing more work, we've got more context switching going on."
We urgently need more public traces of AI agents attacking and defending systems. Defenders can’t learn from what they can’t see. If you have traces and are being pressured to keep them private, my DMs are open. Let’s level the playing field and fight the asymmetry and lack of transparency in AI!
As a researcher who did early work on the productivity impacts of AI chatbots using RCTs, I’d note a lack of similar studies since the dawn of true agents last fall Partially that is newness & partially research design challenges, but I suspect we are missing some large effects.
By the way, could Grok Bot's model already be Opus 5.5? It works remarkably well on complex tasks.
Wall Street is increasingly letting AI agents interpret market information, so a bad source or missing context can spread into real-world decisions before anyone detects the error. The source article does not identify specific models, products, or institutions.
Alibaba's official handbook on AI Native R&D identifies four open challenges: infrastructure engineering complexity, enterprise knowledge assets not yet agent-friendly, organizational design, and the pace of AI iteration. The post's author argues that Agent Infra must suit non-deterministic agent operation and that enterprise knowledge needs top-down structuring and governance. The author also notes that organizational resistance in large companies makes AI adoption harder than in startups.
Researchers at robotics labs say that AI advances behind chatbots like ChatGPT and Claude will not quickly produce robots that are useful in everyday life. Many skeptics argue that using language- and image-based intelligence to master the physical world is far harder than it sounds, despite bold predictions from Elon Musk about Tesla's Optimus. Progress is real but incremental, as shown by Google DeepMind's Gemini Robotics controlling ALOHA 2 arms to pack a lunchbox.
AVEVA chief technologist Arti Garg argues industrial AI should augment rather than replace human supervisors in critical decisions, with guardrails defining where automated systems can act. She says organizations must rethink business processes and safeguards as foundation models, physical AI, and agentic AI enable more complex automation.
The author argues that in agent workflows, the real bottleneck is whether users can precisely express requirements, not the model or agent capability. When people work outside their expertise, they lack the precision needed for prompts and plans, forcing many imprecise iterations that waste time and tokens. The suggested fix is to have the model first teach the unfamiliar domain knowledge before acting.
My Grok bot has found its first powerful use case. Using a personal agent to collect information from Twitter for you is far better than scrolling your phone yourself, manually gathering information, or getting addicted.
Man, I wish Steve Jobs were still alive. He would’ve built the best personal AI agent. The biggest limitation of Instint and Muse is that they can’t control most apps on my phone. Apple controls the entire iOS ecosystem. No company is better positioned to build an AI agent that can do literally everything on your phone. And yet somehow, Siri still sucks…
Block's AI capabilities lead describes using Claude Fable to plan large code migrations and direct smaller models like Opus and Sonnet on individual tasks. He says Block routes frontier and smaller models by task and keeps merges and production deploys behind human dual approval.
AIWhy it matters: Block's engineering lead describes how frontier models orchestrate large migrations and how access, effort levels, and safeguards are managed across an organization.
and because agents make it easier, almost everyone is operating outside their expertise at some point but luckily, you can just ask the model to teach you what you don't know
This Next Token episode discusses Personal Agents, including Dots in Codex, memory and cloud computer permissions, and whether agents should act as assistants or digital twins. The hosts also cover Instinct's booking and business-travel model, hands-on projects built with Opus 5.5, and whether software, games, and hardware could become open source as AI makes rewriting easier.
It was one prompt, claude verified pretty well too! No bugs so far