Agents and Autonomy
AI systems that plan, use tools, take multi-step actions, operate computers, coordinate workflows, or perform tasks with limited human supervision.
The Cloud Is Being Rebuilt Around Agents, Tokens, and Sandboxes
Vercel chief executive Guillermo Rauch used a Stanford MS&E435 seminar to argue that coding agents are expanding the software market rather than merely making developers faster. In his view, AI is widening software creation from professional programmers to business users and autonomous agents, while shifting cloud demand from websites and applications toward deployed agents that need token routing, sandboxes, security, observability and long-running compute.
Scientific Discovery Is Being Rebuilt Around Models, Agents, and Lab Automation
NVIDIA argues that scientific discovery is being reorganized around a new bottleneck: instruments and simulations now generate more biological data than researchers can interpret manually. Its case for the “agentic AI era” is that accelerated computing, open biological models, AI agents, and lab automation are becoming a single discovery stack, compressing simulation timelines and shifting more of the work of reading, design, analysis, and execution onto computational systems while scientists define the questions.
AI Engineering Is Moving From Model Benchmarks to Production Harnesses
Shawn “swyx” Wang argues that AI engineering is shifting from a race over raw model capability to the production systems that make models usable: evals, harnesses, memory, routing, infrastructure and auditability. Drawing on Cognition’s Frontier Code benchmark and his view of the AI Engineer agenda, Wang says the key software frontier is no longer whether agents can pass tests, but whether they can produce maintainable, mergeable code inside real organizations. His broader case is that unstable model access and enterprise constraints make the surrounding system, not the model alone, the durable product boundary.
A $280 Single-Board Computer Can Replace a $28 Monthly Agent VPS
A video test of the Rubik Pi 3 argues that a small single-board computer can be enough to run an always-on AI agent server when the language model itself remains in the cloud. Comparing the Qualcomm Dragonwing-based board with a $28-a-month DigitalOcean VPS running OpenClaw, the presenter found that task times were often shaped less by hardware than by the agent framework’s planning, tool choices and execution path. His conclusion is that the Rubik Pi is not a replacement for fast interactive chatbots, but is viable for asynchronous background agents and could replace his VPS.
World’s Fair 2026 Recasts AI Engineering Around Deployment and Operations
Shawn Wang is pitching AI Engineer World’s Fair 2026 as more than a larger version of the group’s past conferences. In a video update, he argues that the San Francisco event is being redesigned around how AI engineering work is changing: a much larger expo floor, closer links between research and shipping teams, high-volume token operations, AI leadership practice, and vertical deployment in sectors such as finance and healthcare.
RecursiveMAS Lets AI Agents Collaborate Without Translating Through English
Károly Zsolnai-Fehér presents RecursiveMAS, a paper by Xiyuan Yang, Jiaru Zou and coauthors, as an attempt to fix a coordination cost in multi-agent AI systems: agents repeatedly translating internal work into English for one another. The paper’s claim is that agents can instead pass latent numerical representations directly, improving collaboration while cutting token use. Zsolnai-Fehér says the reported gains are substantial on small models, including better math results and far fewer tokens, but frames the work as early research rather than a deployable agent product.
Codex Turns Recorded Workflows Into Reusable Editable Skills
OpenAI presents Record & Replay in Codex as a way to turn a demonstrated recurring workflow into an inspectable, editable skill. In the source example, a user records a YouTube upload process once, and Codex converts the observed steps, defaults and file conventions into a reusable `SKILL.md`. The argument is that repeat work can move from long prompts and remembered preferences to short invocations, with Codex applying the learned workflow to the next relevant task.
Flows Agent Turns Creative Briefs Into Editable AI Production Pipelines
ElevenLabs presents Flows Agent as a conversational assistant for building and revising node-based creative workflows inside ElevenCreative Flows. The company’s case is that a user can describe an ad or other asset in natural language, have the agent assemble the models, prompts, nodes, and connections, then keep the resulting pipeline visible for edits, approvals, and reuse. The demo emphasizes cost controls for credit-heavy generation, node-level revisions through chat, and templates that turn a completed flow into a repeatable production system.
Agents Often Claim Web Access After Being Blocked or Challenged
Rafael Levi of Bright Data argues that many web-dependent agents fail not because they cannot produce answers, but because they report success after web access has broken. In a demo using Bright Data’s Web MCP, Levi shows the same agent failing against sites such as LinkedIn, Instagram, Amazon and TikTok without live access, then producing usable results when given infrastructure for search, scraping, JavaScript rendering and CAPTCHA handling. His broader case is that reliable agents need a real public-web access layer, not prompts that assume the model saw the page.
Hermes Uses a Minimal Agent Loop to Preserve State Across Channels
Alejandro AO’s walkthrough of Hermes presents the agent as a deliberately small always-on system rather than a complex orchestration stack. He argues that Hermes’ usefulness comes from a simple loop that builds context from Markdown files, message history, tools, skills and memory, then preserves state through compression, SQLite transcripts, optional external memory providers, gateway integrations and scheduled cron jobs. The architecture’s central concern is continuity: keeping enough context across channels and time for the agent to behave like a persistent assistant.
Enterprise AI Is Blocked by Context, Not Model Intelligence
Databricks chief executive Ali Ghodsi argues that enterprise AI is constrained less by model intelligence than by access to company context: data, documents, processes and relationships that agents need to operate inside businesses. In a Bloomberg Tech interview with Ed Ludlow, Ghodsi said Databricks is building products such as Genie Ontology and Lakehouse to make that context usable, while adoption in critical workflows remains slowed by security, legal and approval processes. He also declined to confirm reports of a new funding round and said Databricks is not rushing toward an IPO.
Apple’s Revamped Siri May Be Good Enough to Ease Its AI Crisis
Bloomberg’s Mark Gurman argues that Apple’s revamped Siri is not a leap ahead of ChatGPT, Gemini or Claude, but may be good enough to stabilize Apple’s position in AI. Speaking with Ed Ludlow, Gurman said the new Siri finally delivers on much of the assistant promise Apple made years ago, while still falling short on advanced tasks such as deep research, long-document summaries and creating spreadsheets or slide decks. His case is that Apple can ease its AI crisis if Siri now handles the everyday questions and device-assistant tasks most of its 2bn-plus users actually need.
Tokens Can Now Substitute for 100-Person Startup Engineering Teams
In a Stanford CS153 lecture, OpenAI chief executive Sam Altman argued that AI has already rewritten the startup playbook, allowing small teams to buy capabilities with tokens that once required large engineering organizations. He used OpenAI’s experience with ChatGPT, Codex and model scaling to make a broader case: scale keeps producing capabilities that experts underestimate, but the institutions around AI — from education and research pipelines to compute markets and governance — are not adapting as quickly. Altman said the central choice ahead is whether intelligence becomes a broadly available utility or remains concentrated in a few companies.
AI Market Power Is Moving Beyond the Frontier Model
Alex Kantrowitz and Ranjan Roy argue that the AI market is shifting away from standalone model capability and toward control of infrastructure, access and workflow layers. Their discussion frames SpaceX’s IPO as a public-market AI-cloud story that complicates OpenAI’s ambitions, Anthropic’s Fable rollout as a case where safety policy also looks like market power, and OpenAI’s possible price cuts as a test of whether frontier models can remain premium products. Apple’s Siri, in their telling, matters for the same reason: usefulness may come less from the best model than from where the model sits.
Codex Turns Customer Reviews Into Website Mockups for Sales Demos
OpenAI solutions engineer Stephanie Anani presents Codex as a practical partner for solutions engineering, not just a coding tool. Her example starts with a customer’s Trustpilot reviews, uses Codex to analyze what end users are saying, and then turns that feedback into a website mockup that shows the customer how changes could look in its own context. Anani’s case is that Codex is most useful when it works inside a user’s existing materials and workflows, including by preserving strong outputs as reusable skills.
Anthropic’s Fable Backlash Exposes the Risk of Hidden AI Gatekeeping
The All-In panel argues that Anthropic’s handling of Claude Fable 5 turned AI safety into an enterprise trust problem, with Jason Calacanis, Chamath Palihapitiya, David Sacks and David Friedberg focusing on hidden downgrades, prompt retention and a provider’s power to decide who receives full model capability. The same concern over opaque discretion shaped their California election discussion, where Friedberg and Sacks argued that legal ballot rules can still produce outcomes voters view as manipulated, while Calacanis called for investigation rather than treating suspicious statistics as proof of fraud.
SpaceX’s IPO Forces Public Markets to Price a Venture-Scale Future
Jason Calacanis used SpaceX’s reported IPO to argue that public markets will misread the company if they treat it only as a near-term earnings story. On This Week in Startups, he framed SpaceX as part operating business and part venture bet: Starlink and launch can be measured today, while direct-to-phone service, orbital data centers, Moon bases and Mars remain longer-horizon wagers on Elon Musk’s execution. The episode then turned to Polsia founder Ben Cera, whose AI-run fundraising stunt was presented as a case study in attention that demonstrates the product rather than merely promoting it.
Codex Turns Earnings Reports Into Post-Quarter Investment Thesis Updates
OpenAI is pitching Codex’s public-equity investing plugin as a way to turn a company’s latest quarter into thesis-revision work rather than a conventional earnings recap. Using a Cava post-earnings example, the source argues that Codex can combine first-party filings, earnings-call material and third-party data from sources including Quartr, Daloopa and S&P Global to separate business momentum from stock expectations, build bull, base and bear cases, and produce a monitoring checklist for the next reporting window.
AI’s Economic Test Is Broad Diffusion, Not Frontier Capability
Microsoft chief executive Satya Nadella told a New York Times Hard Fork live audience that AI’s economic test is not whether a few companies build stronger frontier models, but whether the technology spreads widely enough to raise productivity, justify its token costs and create visible benefits for workers and communities. He argued that Microsoft’s role is to build platforms for that diffusion, while warning that job displacement, data center burdens and concentrated gains will make the backlash rational unless humans remain stakeholders through new “glue work” and local upside.
Codex Adds Chrome DevTools Access for Web App Debugging
OpenAI says Codex’s Browser Use can now connect to the Chrome DevTools Protocol, allowing it to inspect running web applications through console logs, runtime errors, local storage, styling, network traffic and performance profiles. The source argues that this moves Codex debugging beyond code inspection: in a slow chat-app example, Codex profiles interactions, identifies duplicate requests and expensive server paths, makes targeted fixes, and reports before-and-after timings. The capability is gated behind Developer mode and per-site approval because CDP access can expose sensitive browser internals.
Codex Turns Salesforce Account Context Into Seller-Ready Prospecting Work
OpenAI’s demo presents Codex as a workflow layer for sales prospecting, connecting Salesforce, company sales templates and Gmail to turn account context into seller-ready work. The sales plugin is shown prioritizing accounts, generating a standardized pursuit plan, drafting account-specific outreach in Gmail and setting up a governed morning cadence that updates the plan and prepares follow-up drafts without sending them automatically.
Human Attention Is Becoming the Bottleneck in AI Coding Workflows
Zack Proser, an Applied AI engineer at WorkOS, argues that AI coding has shifted the bottleneck from tool speed to human attention. His proposed workflow uses voice dispatch, isolated git worktrees, Slack and Linear-reading agents, remote phone control, and layered verification so developers can keep agent loops moving without staying pinned to a desk or rubber-stamping work they can no longer track.
Models Will Absorb Today’s Agent Harnesses Within a Year
Logan Kilpatrick, who leads Google AI Studio and the Gemini API, argues that the current rush to build agent harnesses may have a short shelf life. In an interview with Sequoia Capital’s Sonya Huang, he says models are absorbing the scaffolding around agents and could make much of today’s custom harness layer less distinctive within about 12 months. Google’s own strategy runs on both sides of that claim: Antigravity has become a shared agent layer across products, while Kilpatrick says the durable advantage for builders will move to focus, domain knowledge, risk tolerance and useful outcomes for users.
Affirm’s Founder Says Consumer Finance Should Not Profit From Confusion
Max Levchin, the PayPal co-founder and Affirm chief executive, tells Tim Ferriss that his career has been shaped by a preference for confronting constraints directly rather than explaining them away. Across PayPal, his childhood in the Soviet Union, and Affirm’s design, Levchin argues that technically elegant systems fail when they ignore human behavior, bad incentives, or user experience. His case is that better companies and decisions come from making the real trade-offs visible, whether in leadership, consumer credit, AI commerce, or personal discipline.