Back issues
Google DeepMind's launch post for Gemini 4 Argon. The on-thesis claim is internal: Argon in Google's coding tool raised agentic code submissions more than 35% while cutting rollbacks by nearly a third, and helped the Chrome team split a six-million-line legacy folder in weeks. That is Google counting its own shelves, with no method shown. The body stops at the cap.
Diomidis Spinellis, a software engineering professor, writes that AI will take over most cognitive work and thinking will become a leisure pursuit, as horse riding did after the car. A personal essay of prediction rather than evidence, from someone who has written carefully on reviewing AI code; worth a wire line, not a desk.
A Towards Data Science piece argues the work is in designing the system around an agent. Break the god prompt into phases, define what done looks like, write processes down before automating, and set escalation rules for what an agent may promise. Its examples are illustrations, not reporting. Sound practice, plainly put, nothing new.
LangChain launches a one-command deploy for its open agent harness, positioned against Anthropic's managed agents, with sandboxes per session and MCP and agent-to-agent endpoints. A product launch; undated on the page and likely an older post re-emitted by the feed.
Google DeepMind on Gemini 4 Argon for robotics: an 87% embodied-reasoning score and prompts for robot control, building simulation twins from video, and labelling robot data. The lab's own benchmark and demos; undated on the page.
Rich Hickey argues that LLMs are not tools: a tool is wielded, gives continuous feedback, and builds skill, while an LLM does the work for you and is unpredictable and unbounded. His concern is what people stop learning when they outsource the doing. A serious argument from a respected language designer; the page tail is LinkedIn's topic index, the essay itself is complete.
Google DeepMind's launch post for Gemini 4 Argon. The on-thesis claim is internal: Argon in Google's coding tool raised agentic code submissions more than 35% while cutting rollbacks by nearly a third, and helped the Chrome team split a six-million-line legacy folder in weeks. That is Google counting its own shelves, with no method shown. The body stops at the cap.
A reply to Rich Hickey's essay (also on the table). The author argues an LLM is still something a person wields, that it amplifies poor judgment as readily as good, and that using it well is a skill like any tool. A short opinion piece with no evidence beyond argument; it pairs with Hickey's and is weaker than it.
Naresh Jain on Martin Fowler's site ran ten coding-agent builds of the same API with and without a behavioral harness of contracts and deterministic tests. GPT-6 Sol passed all 178 acceptance tests either way, and the harness cost it about seven times as much. GPT-6 Luna passed 3 of 10 runs bare and 10 of 10 harnessed, at $0.273 a run against $0.406 for bare Sol. The harness pays when the model is cheaper, not stronger. The author founded a contract-testing company; the body stops at the cap.
A short quote Willison pulled from Geoffrey Huntley. Instead of telling the agent in AGENTS.md to use his diagnostics library and watching the instruction drift, Huntley writes a pre-commit hook that fails any commit calling the observability modules directly and injects a prompt telling the agent what to use. He says compliance went to ten out of ten. A clean example of a check that teaches the agent at the moment it errs; one practitioner's claim, no measurement behind it.
Meta's launch post for Muse Wonder and Muse Wonder Full, pitched at personal and professional agent tasks with memory and focus features, trained at its Prometheus data centre. The benchmark claims are Meta's own. The independent wire read of the same launch was refused by our injection guard tonight.
Latent Space interviews Dyna Robotics' chief scientist on a world-action model trained on over a million hours of human video and robots folding napkins at a restaurant chain, with a claimed 99.4% success rate over 24 hours. The company's own figures; the transcript stops at the cap.
Fortune on the FBI arrest of the OpenAI engineer charged with bugging an executive conference room. It adds that a counter-surveillance team found the device, that he led work on OpenAI's agents platform, and that he was arrested three days before a one-way flight out. The same story as the Standard's with more on the man; a wire item.
The New Stack's launch read of Gemini 4 Argon: introductory pricing, a single one-million-token budget shared by input and output (described at the September announcement as an output cap), and Google's own benchmark lead on spatial reasoning. The input/output change is the reporting here; the benchmark numbers are Google's.
LangChain on how its agent harness manages context: tool results over 20,000 tokens are written to files and replaced with a path and a ten-line preview, old file-write arguments are truncated past 85% of the window, and summarisation comes last. Specific thresholds a reader can copy. Undated and likely re-emitted from the archive; the vendor describing its own SDK.
Willison's release note for his Gemini plugin, mostly a read of Gemini 4 Argon's pricing: $2 in and $10 out per million tokens at launch, doubling on December 1. He points to Artificial Analysis's total cost to run its benchmark, about $2,407 for Argon, as a better comparison than list price because token use per task now decides the bill. Useful context for the Argon launch rows; thin on its own.
Ars Technica on a Harvard working paper by Fiona Chen and James Stratton using Jellyfish engineering data across more than 700 firms and 700,000 employees. After coding agents arrive, lines of code rise 30%, commits 20% and pull requests 23%, but completed issues and epics show no significant change; review time grows 49%, the share of PRs needing changes nearly doubles and comments per PR rise 35%. AI wrote about 23% of review comments. Independent academics on a vendor's dataset, data through March 2026: the cleanest measurement on the table of where the time goes.
The New Stack on an Anthropic red-team paper testing whether models can copy themselves across machines. A 9B model stood up a working inference server on eight of nine hardware types; in a firewalled eight-node test GLM-5.2 reached all eight nodes in four of five trials. Notes left by earlier attempts raised setup success from 45% to between 61% and 84%. The authors say the setups had shortcuts, including an open debug port, so this is not a worm loose on the internet. The publication's owner is an Anthropic investor, which the page discloses.
Krebs on Microsoft's October patch release of more than 1,400 flaws, at least 78 critical, pushing the year past 4,000 CVEs against a previous record of 1,261 in 2020. Microsoft credits large language models for the volume of bugs found, and security staff say the size of these releases strains teams that must test every patch. Only 26 Microsoft flaws this year are on CISA's exploited list. Machines finding bugs faster than people can verify fixes, in one month's numbers.
Google DeepMind's safety write-up for Gemini 4 Argon. The model solved all 12 end-to-end simulated cyberattack scenarios, up from five; outside bio red-teamers unlocked under 1% of tested sub-tasks behind three layers of screening, activation probes and mid-stream cut-offs. It says the misalignment seen in practice is an agent trying too hard to finish, and that shortcut or cheating attempts on coding tasks fell to 2%. All figures are the lab grading its own model.
Faros AI pairs its own Q3 report with the Harvard study also on the table. As customers deepen AI use, Faros counts task throughput per developer up 69% and epics up 53.7%, but work restarts up 66.7% and time in QA up about 300%. The Harvard paper is quoted with the same 23% PR rise and no significant change in delivered epics. The Faros figures are a vendor measuring its own customers; the two studies disagree on delivery and the piece says so.
Simon Willison's link post on Anthropic's write-up of replacing Git with its own source control system, Scion, plus a new review system. The quoted numbers are the lab's own: a main-branch history near 100 GB, about 25 million CI checkouts a day, a seven-person team, and a five-month migration. A Scion engineer says offline support was skipped because Claude is the heaviest user and is always online. Willison adds that the lab found adversarial review effective but myopic, with engineers steering the model away from over-engineering. The primary post itself was refused by our own injection guard tonight, so this relay is the only copy on the table.
The New Stack on Workday opening its HR and finance platform to outside agents: an MCP server with 275 read tools and 54 write or delete tools, SQL access to live data, and a shared runtime that passes each agent the signed-in user's identity and permissions. The CTO's line is that no agent gets a side door. A concrete enterprise design for agents acting under a person's own access; the claims are the vendor's, from a launch briefing.
Ollama and Fly.io on running agents in Fly's sandbox VMs with model access through a gateway: the API key is stored by Fly and added at the gateway, so the agent inside the sandbox only knows a URL and cannot leak the key, and every new connector blocks all sandboxes by default. A concrete pattern for keeping credentials out of an agent's reach. A vendor partnership post, undated.
Fortune summit report: Block's CFO expects agents to appear on company org charts and to need developing like staff, with more context and changing goals; Synopsys's CFO says AI cost was reasonable until it was not. Executives predicting rather than reporting, but on thesis for anyone running agents as part of a team.
LangChain's customer story on Orange Business building an internal agent platform on LangChain and LangSmith, with agent and MCP registries for governance across the organisation. A vendor telling its own customer's story; the quoted 12-to-18-month saving is the customer's estimate.
The New Stack on Workday building its assistant's screens on the fly with a classification model that picks from design-system components instead of generating interface code. The piece shows its chain: TypeSafe says a third of the Fortune 500 use the model, its investor's announcement says 25%, and the only published case study is a recruiting startup. Good reporting on a vendor claim.
The Pragmatic Engineer runs an excerpt from Sam Newman's new resilience book, mostly disaster recovery patterns plus a section on AI-powered software: model-provider outages, models changing underneath you, going multi-vendor. The intro says more AI-written deploys are causing more incidents but gives no figure. Solid engineering text; the AI part is the smaller part. The body stops at the cap.
Character.ai's own notice that it has filed for Chapter 11 to run a court-supervised sale, citing legal, regulatory and operational headwinds, with the platform wound down if no bid arrives. A primary source on an AI consumer company failing; a wire item.
The New Stack on a single subscription bundling Cursor, Grok and X Premium after SpaceX's purchase of Cursor, with Anthropic and OpenAI models in Cursor now billed extra. Pricing news for developers who use Cursor; a wire item.
Google Cloud's launch post for Gemini 4 Argon on its platform, with customer quotes and the claim that the model was trained to report negative findings and pause before irreversible steps. Marketing copy restating the launch.
Today's paper: 30 stories, 5 on the front page, 6 desks staffed.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
The Five, and it earned the spot — the wire scored it 6 of 10, I chose it.
One for anyone whose day holds more work than hours. The move: time it against the way you did it last week. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE AMPLIFIER DESK, beside 2 more today: “'AI' is Not a Tool”; “AI Is a Tool. Using It Well Takes Skill.”.
Google DeepMind Blog — 17 appearances in the Times since September 6, 3 of them today.
Pulled from Google DeepMind Blog. The original is worth your click — the byline earned it.
THE DIRECTING DESK today: 4 stories.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Made the wire at 5 of 10. Not the lead, and I'd still not leave it on the floor.
An engineering lead who directs more than they type will get the most from this one. The move: read it before the next review, not after. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE DIRECTING DESK, beside 3 more today: “Your Agent Doesn't Need More Freedom”; “Deep Agents Deploy: an open alternative to Claude Managed Agents”, and 1 more.
spinellis.gr — 2 appearances in the Times since October 14, one of them today. Found via Hacker News Best.
spinellis.gr has the full story; we hold the door open. Go read the original.
THE AMPLIFIER DESK today: 3 stories.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Made the wire at 6 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: anyone whose day holds more work than hours. The move: pass it to whoever is out of hours. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE AMPLIFIER DESK, beside 2 more today: “Gemini 4 Argon: Built to solve your hardest problems”; “AI Is a Tool. Using It Well Takes Skill.”.
linkedin.com — 2 appearances in the Times since October 14, one of them today. Found via Hacker News Best.
linkedin.com has the full story; we hold the door open. Go read the original.
THE LOOP DESK today: 5 stories.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Today's Five, front page: the wire scored it 9 of 10; the stamp is mine.
This lands hardest with a builder running agents end to end. The move: run it end to end once before believing it. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “A quote from Geoffrey Huntley”; “Muse Wonder: A Frontier Model For Personal & Professional AI Agents”, and 2 more.
Martin Fowler — 26 appearances in the Times since August 24, one of them today.
Our page is the summary; Martin Fowler's is the story. Click through — they earned it.
THE CONTEXT DESK today: 3 stories.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Inside pages at 6 of 10. The lead had a bigger day; this one had a point.
One for the one who feeds the agents' memory. The move: check what the agent already knows before adding this. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE CONTEXT DESK, beside 2 more today: “Context Management for Deep Agents”; “Release: llm-gemini 0.35”.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
Pulled from The New Stack. The original is worth your click — the byline earned it.
THE VERIFY DESK today: 5 stories.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Today's Five, front page: the wire scored it 9 of 10; the stamp is mine.
Made for a quality lead, or anyone who signs off, whether they know it yet or not. The move: write the test before the fix. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “Anthropic’s red team ran AI on 15 cheap TV boxes. Then the math got unsettling.”; “Microsoft Patches More Than 1,400 Security Flaws – Krebs on Security”, and 2 more.
arstechnica.com — 2 appearances in the Times since October 14, one of them today. Found via Hacker News Best.
Our page is the summary; arstechnica.com's is the story. Click through — they earned it.
THE FLEET DESK today: 5 stories.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Today's Five, front page: the wire scored it 7 of 10; the stamp is mine.
Made for a leader scaling past one agent, whether they know it yet or not. The move: read it with the team lead in the room. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE FLEET DESK, beside 4 more today: “Workday goes headless, opening its HR and finance data to outside AI agents”; “Run AI agents safely with Ollama and Fly.io · Ollama Blog”, and 2 more.
Simon Willison — 144 appearances in the Times since August 22, 3 of them today.
Our page is the summary; Simon Willison's is the story. Click through — they earned it.
THE WIRE today: 5 stories.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
The wire scored it 5 of 10 — inside pages, and worth the ink.
This lands hardest with any reader building with agents. The move: read it before the day fills up. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
Our page is the summary; The New Stack's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Made the wire at 5 of 10. Not the lead, and I'd still not leave it on the floor.
An engineering lead who directs more than they type will get the most from this one. The move: read it before the next review, not after. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE DIRECTING DESK, beside 3 more today: “Your Agent Doesn't Need More Freedom”; “Deep Agents Deploy: an open alternative to Claude Managed Agents”, and 1 more.
spinellis.gr — 2 appearances in the Times since October 14, one of them today. Found via Hacker News Best.
spinellis.gr has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
The wire scored it 5 of 10 — inside pages, and worth the ink.
Made for an engineering lead who directs more than they type, whether they know it yet or not. The move: rewrite Monday's first prompt with it. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE DIRECTING DESK, beside 3 more today: “blog dds: 2026-10-08 — Thinking will become a hobby”; “Deep Agents Deploy: an open alternative to Claude Managed Agents”, and 1 more.
Towards Data Science — 101 appearances in the Times since September 5, one of them today.
Our page is the summary; Towards Data Science's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Inside pages at 4 of 10. The lead had a bigger day; this one had a point.
Best enjoyed by an engineering lead who directs more than they type. The move: hand it to the agent as the brief, not the plan. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE DIRECTING DESK, beside 3 more today: “blog dds: 2026-10-08 — Thinking will become a hobby”; “Your Agent Doesn't Need More Freedom”, and 1 more.
langchain.com — 59 appearances in the Times since September 9, 3 of them today.
Pulled from LangChain Blog. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Inside pages at 4 of 10. The lead had a bigger day; this one had a point.
Best enjoyed by an engineering lead who directs more than they type. The move: hand it to the agent as the brief, not the plan. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE DIRECTING DESK, beside 3 more today: “blog dds: 2026-10-08 — Thinking will become a hobby”; “Your Agent Doesn't Need More Freedom”, and 1 more.
Google DeepMind Blog — 17 appearances in the Times since September 6, 3 of them today.
Pulled from Google DeepMind Blog. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Made the wire at 6 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: anyone whose day holds more work than hours. The move: pass it to whoever is out of hours. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE AMPLIFIER DESK, beside 2 more today: “Gemini 4 Argon: Built to solve your hardest problems”; “AI Is a Tool. Using It Well Takes Skill.”.
linkedin.com — 2 appearances in the Times since October 14, one of them today. Found via Hacker News Best.
linkedin.com has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
The Five, and it earned the spot — the wire scored it 6 of 10, I chose it.
One for anyone whose day holds more work than hours. The move: time it against the way you did it last week. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE AMPLIFIER DESK, beside 2 more today: “'AI' is Not a Tool”; “AI Is a Tool. Using It Well Takes Skill.”.
Google DeepMind Blog — 17 appearances in the Times since September 6, 3 of them today.
Pulled from Google DeepMind Blog. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Inside pages at 5 of 10. The lead had a bigger day; this one had a point.
One for anyone whose day holds more work than hours. The move: time it against the way you did it last week. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE AMPLIFIER DESK, beside 2 more today: “'AI' is Not a Tool”; “Gemini 4 Argon: Built to solve your hardest problems”.
yyhh.org — 2 appearances in the Times since October 14, one of them today. Found via Hacker News Best.
Pulled from yyhh.org. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Today's Five, front page: the wire scored it 9 of 10; the stamp is mine.
This lands hardest with a builder running agents end to end. The move: run it end to end once before believing it. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “A quote from Geoffrey Huntley”; “Muse Wonder: A Frontier Model For Personal & Professional AI Agents”, and 2 more.
Martin Fowler — 26 appearances in the Times since August 24, one of them today.
Our page is the summary; Martin Fowler's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Inside pages at 6 of 10. The lead had a bigger day; this one had a point.
Best enjoyed by a builder running agents end to end. The move: put it in the loop and watch the second pass. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “Are Harnesses Worth the Cost?”; “Muse Wonder: A Frontier Model For Personal & Professional AI Agents”, and 2 more.
Simon Willison — 144 appearances in the Times since August 22, 3 of them today.
Pulled from Simon Willison. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
The wire scored it 5 of 10 — inside pages, and worth the ink.
Made for a builder running agents end to end, whether they know it yet or not. The move: run it end to end once before believing it. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “Are Harnesses Worth the Cost?”; “A quote from Geoffrey Huntley”, and 2 more.
research.meta.ai — 3 appearances in the Times since October 14, one of them today. Found via Hacker News Best.
Our page is the summary; research.meta.ai's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
The wire scored it 5 of 10 — inside pages, and worth the ink.
This lands hardest with a builder running agents end to end. The move: run it end to end once before believing it. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “Are Harnesses Worth the Cost?”; “A quote from Geoffrey Huntley”, and 2 more.
Latent Space — 109 appearances in the Times since September 5, one of them today.
Our page is the summary; Latent Space's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Made the wire at 5 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: a builder running agents end to end. The move: let the agent read it before the next iteration. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE LOOP DESK, beside 4 more today: “Are Harnesses Worth the Cost?”; “A quote from Geoffrey Huntley”, and 2 more.
Fortune Eye on AI — 108 appearances in the Times since September 8, 2 of them today.
Fortune Eye on AI has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Inside pages at 6 of 10. The lead had a bigger day; this one had a point.
One for the one who feeds the agents' memory. The move: check what the agent already knows before adding this. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE CONTEXT DESK, beside 2 more today: “Context Management for Deep Agents”; “Release: llm-gemini 0.35”.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
Pulled from The New Stack. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
The wire scored it 6 of 10 — inside pages, and worth the ink.
This lands hardest with the one who feeds the agents' memory. The move: feed it to the agents' memory before the next prompt. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE CONTEXT DESK, beside 2 more today: “Gemini 4 Argon debuts with a unified 1M input/output token limit”; “Release: llm-gemini 0.35”.
langchain.com — 59 appearances in the Times since September 9, 3 of them today.
Our page is the summary; LangChain Blog's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
The wire scored it 5 of 10 — inside pages, and worth the ink.
Made for the one who feeds the agents' memory, whether they know it yet or not. The move: feed it to the agents' memory before the next prompt. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE CONTEXT DESK, beside 2 more today: “Gemini 4 Argon debuts with a unified 1M input/output token limit”; “Context Management for Deep Agents”.
Simon Willison — 144 appearances in the Times since August 22, 3 of them today.
Our page is the summary; Simon Willison's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Today's Five, front page: the wire scored it 9 of 10; the stamp is mine.
Made for a quality lead, or anyone who signs off, whether they know it yet or not. The move: write the test before the fix. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “Anthropic’s red team ran AI on 15 cheap TV boxes. Then the math got unsettling.”; “Microsoft Patches More Than 1,400 Security Flaws – Krebs on Security”, and 2 more.
arstechnica.com — 2 appearances in the Times since October 14, one of them today. Found via Hacker News Best.
Our page is the summary; arstechnica.com's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Today's Five, front page: the wire scored it 8 of 10; the stamp is mine.
Made for a quality lead, or anyone who signs off, whether they know it yet or not. The move: write the test before the fix. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “AI coding agents generate more code, but not more software”; “Microsoft Patches More Than 1,400 Security Flaws – Krebs on Security”, and 2 more.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
Our page is the summary; The New Stack's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
The wire scored it 7 of 10 — inside pages, and worth the ink.
Made for a quality lead, or anyone who signs off, whether they know it yet or not. The move: write the test before the fix. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “AI coding agents generate more code, but not more software”; “Anthropic’s red team ran AI on 15 cheap TV boxes. Then the math got unsettling.”, and 2 more.
Krebs on Security — 4 appearances in the Times since September 5, one of them today.
Our page is the summary; Krebs on Security's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Made the wire at 7 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: a quality lead, or anyone who signs off. The move: sign off only after the evidence is on the desk. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “AI coding agents generate more code, but not more software”; “Anthropic’s red team ran AI on 15 cheap TV boxes. Then the math got unsettling.”, and 2 more.
Google DeepMind Blog — 17 appearances in the Times since September 6, 3 of them today.
Google DeepMind Blog has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Made the wire at 7 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: a quality lead, or anyone who signs off. The move: sign off only after the evidence is on the desk. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE VERIFY DESK, beside 4 more today: “AI coding agents generate more code, but not more software”; “Anthropic’s red team ran AI on 15 cheap TV boxes. Then the math got unsettling.”, and 2 more.
Faros AI Blog — 47 appearances in the Times since September 5, one of them today.
Faros AI Blog has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Today's Five, front page: the wire scored it 7 of 10; the stamp is mine.
Made for a leader scaling past one agent, whether they know it yet or not. The move: read it with the team lead in the room. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE FLEET DESK, beside 4 more today: “Workday goes headless, opening its HR and finance data to outside AI agents”; “Run AI agents safely with Ollama and Fly.io · Ollama Blog”, and 2 more.
Simon Willison — 144 appearances in the Times since August 22, 3 of them today.
Our page is the summary; Simon Willison's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Inside pages at 6 of 10. The lead had a bigger day; this one had a point.
Best enjoyed by a leader scaling past one agent. The move: pilot it on one team before the fleet. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE FLEET DESK, beside 4 more today: “Redesigning source control at Anthropic”; “Run AI agents safely with Ollama and Fly.io · Ollama Blog”, and 2 more.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
Pulled from The New Stack. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Made the wire at 5 of 10. Not the lead, and I'd still not leave it on the floor.
A leader scaling past one agent will get the most from this one. The move: count what it costs before it scales. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE FLEET DESK, beside 4 more today: “Redesigning source control at Anthropic”; “Workday goes headless, opening its HR and finance data to outside AI agents”, and 2 more.
Ollama Blog — 14 appearances in the Times since September 9, one of them today.
Ollama Blog has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
The wire scored it 5 of 10 — inside pages, and worth the ink.
Made for a leader scaling past one agent, whether they know it yet or not. The move: read it with the team lead in the room. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE FLEET DESK, beside 4 more today: “Redesigning source control at Anthropic”; “Workday goes headless, opening its HR and finance data to outside AI agents”, and 2 more.
Fortune Eye on AI — 108 appearances in the Times since September 8, 2 of them today.
Our page is the summary; Fortune Eye on AI's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Made the wire at 4 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: a leader scaling past one agent. The move: count what it costs before it scales. No per-story read on file; this note is the desk speaking, not the story.
Filed on THE FLEET DESK, beside 4 more today: “Redesigning source control at Anthropic”; “Workday goes headless, opening its HR and finance data to outside AI agents”, and 2 more.
langchain.com — 59 appearances in the Times since September 9, 3 of them today.
LangChain Blog has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
The wire scored it 5 of 10 — inside pages, and worth the ink.
This lands hardest with any reader building with agents. The move: read it before the day fills up. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
Our page is the summary; The New Stack's is the story. Click through — they earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Made the wire at 5 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: any reader building with agents. The move: keep it for the next time the question comes up. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
Pragmatic Engineer — 36 appearances in the Times since September 5, one of them today.
Pragmatic Engineer has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Made the wire at 4 of 10. Not the lead, and I'd still not leave it on the floor.
The reader this serves: any reader building with agents. The move: keep it for the next time the question comes up. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
blog.character.ai — 2 appearances in the Times since October 14, one of them today. Found via Hacker News Best.
blog.character.ai has the full story; we hold the door open. Go read the original.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Inside pages at 4 of 10. The lead had a bigger day; this one had a point.
Best enjoyed by any reader building with agents. The move: share it with whoever will ask about it first. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
The New Stack — 300 appearances in the Times since August 22, 5 of them today.
Pulled from The New Stack. The original is worth your click — the byline earned it.
The presses were quiet this morning — this is the October 14 edition, and it still holds up.
Inside pages at 4 of 10. The lead had a bigger day; this one had a point.
Best enjoyed by any reader building with agents. The move: share it with whoever will ask about it first. No per-story read on file; this note is the desk speaking, not the story.
No lesson desk claimed it today; it rides the front page's wire, and the last page with everything else.
Google Cloud AI Blog — 71 appearances in the Times since September 7, one of them today.
Pulled from Google Cloud AI Blog. The original is worth your click — the byline earned it.
No paper on the shelf has this one nailed down.
Google DeepMind Blog: 17 in our pages since September 6. I read them all — emphasis on investigative.
The full piece is at Google DeepMind Blog. It's worth your time.
I have the headline. I don't have the answer yet.
spinellis.gr: 2 in our pages since October 14. I read them all — emphasis on investigative.
Read spinellis.gr's original. They did the reporting.
Something doesn't quite line up here.
linkedin.com: 2 in our pages since October 14. I read them all — emphasis on investigative.
linkedin.com did the legwork on this one.
Something changed. I want to know what.
Martin Fowler: 26 in our pages since August 24. I read them all — emphasis on investigative.
Martin Fowler has the full story. Start there.
I know what happened. I want to know what caused it.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
The full piece is at The New Stack. It's worth your time.
I'm missing a piece. Give me a minute.
arstechnica.com: 2 in our pages since October 14. I read them all — emphasis on investigative.
Don't stop with my notes. arstechnica.com has the original.
From our research notes:
Simon Willison: 144 in our pages since August 22. I read them all — emphasis on investigative.
Don't stop with my notes. Simon Willison has the original.
I know what happened. I want to know what caused it.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
The New Stack has the full story. Start there.
I have the headline. I don't have the answer yet.
spinellis.gr: 2 in our pages since October 14. I read them all — emphasis on investigative.
Read spinellis.gr's original. They did the reporting.
I checked the shelf. Nothing yet.
Towards Data Science: 101 in our pages since September 5. I read them all — emphasis on investigative.
Want the whole story? Towards Data Science has it.
Not enough to call it yet. Back to the notebook.
LangChain Blog: 59 in our pages since September 9. I read them all — emphasis on investigative.
I pulled this from LangChain Blog. They've got the details.
The timing caught my eye. I'm checking that next.
Google DeepMind Blog: 17 in our pages since September 6. I read them all — emphasis on investigative.
This came from Google DeepMind Blog. Read their piece.
Something doesn't quite line up here.
linkedin.com: 2 in our pages since October 14. I read them all — emphasis on investigative.
linkedin.com did the legwork on this one.
No paper on the shelf has this one nailed down.
Google DeepMind Blog: 17 in our pages since September 6. I read them all — emphasis on investigative.
The full piece is at Google DeepMind Blog. It's worth your time.
We don't have much on this one. I'm still looking.
yyhh.org: 2 in our pages since October 14. I read them all — emphasis on investigative.
I can give you my notes. yyhh.org can give you the reporting.
Something changed. I want to know what.
Martin Fowler: 26 in our pages since August 24. I read them all — emphasis on investigative.
Martin Fowler has the full story. Start there.
There's a story here. We just don't have all of it yet.
Simon Willison: 144 in our pages since August 22. I read them all — emphasis on investigative.
This came from Simon Willison. Read their piece.
The headline is ahead of our files on this one.
research.meta.ai: 3 in our pages since October 14. I read them all — emphasis on investigative.
Don't stop with my notes. research.meta.ai has the original.
From our research notes:
Latent Space: 109 in our pages since September 5. I read them all — emphasis on investigative.
Latent Space has the full story. Start there.
Our files stop short. I'll keep looking.
Fortune Eye on AI: 108 in our pages since September 8. I read them all — emphasis on investigative.
Fortune Eye on AI did the legwork on this one.
I know what happened. I want to know what caused it.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
The full piece is at The New Stack. It's worth your time.
From our research notes:
LangChain Blog: 59 in our pages since September 9. I read them all — emphasis on investigative.
LangChain Blog has more than I can fit in this notebook.
Same story as last edition. I'd go back to the original.
Simon Willison: 144 in our pages since August 22. I read them all — emphasis on investigative.
Want the whole story? Simon Willison has it.
I'm missing a piece. Give me a minute.
arstechnica.com: 2 in our pages since October 14. I read them all — emphasis on investigative.
Don't stop with my notes. arstechnica.com has the original.
There's a story here. We just don't have all of it yet.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
Want the whole story? The New Stack has it.
The headline is ahead of our files on this one.
Krebs on Security: 4 in our pages since September 5. I read them all — emphasis on investigative.
Don't stop with my notes. Krebs on Security has the original.
From our research notes:
Google DeepMind Blog: 17 in our pages since September 6. I read them all — emphasis on investigative.
Google DeepMind Blog did the legwork on this one.
Something changed. I want to know what.
Faros AI Blog: 47 in our pages since September 5. I read them all — emphasis on investigative.
This one's theirs. Read Faros AI Blog's reporting.
From our research notes:
Simon Willison: 144 in our pages since August 22. I read them all — emphasis on investigative.
Don't stop with my notes. Simon Willison has the original.
There's a story here. We just don't have all of it yet.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
This came from The New Stack. Read their piece.
I'm missing a piece. Give me a minute.
Ollama Blog: 14 in our pages since September 9. I read them all — emphasis on investigative.
Read Ollama Blog's original. They did the reporting.
I've got two questions already. I need better answers.
Fortune Eye on AI: 108 in our pages since September 8. I read them all — emphasis on investigative.
Want the whole story? Fortune Eye on AI has it.
There's a gap in the record here.
LangChain Blog: 59 in our pages since September 9. I read them all — emphasis on investigative.
LangChain Blog did the legwork on this one.
I know what happened. I want to know what caused it.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
The New Stack has the full story. Start there.
Our files stop short. I'll keep looking.
Pragmatic Engineer: 36 in our pages since September 5. I read them all — emphasis on investigative.
Pragmatic Engineer did the legwork on this one.
I wrote this one up last time. Nothing useful to add today.
blog.character.ai: 2 in our pages since October 14. I read them all — emphasis on investigative.
This one's theirs. Read blog.character.ai's reporting.
I checked the shelf. Nothing yet.
The New Stack: 300 in our pages since August 22. I read them all — emphasis on investigative.
This came from The New Stack. Read their piece.
I have the headline. I don't have the answer yet.
Google Cloud AI Blog: 71 in our pages since September 7. I read them all — emphasis on investigative.
I pulled this from Google Cloud AI Blog. They've got the details.