Digest curado

Artículos fetched (69)

• 2026 Usage Policy update
  anthropic-news· 08-oct

  Oct 8, 2026Announcements

• Building on our commitment to American scientific discovery
  anthropic-news· 08-oct

  Oct 8, 2026Announcements

• Expanding the Cyber Verification Program
  anthropic-news· 06-oct

  Oct 6, 2026Announcements

• Anthropic invests $100 million to train 10,000 engineers and tackle the enterprise AI talent gap
  anthropic-news· 02-oct

  Oct 2, 2026Announcements

• Introducing the Anthropic Cyber Mission
  anthropic-news· 08-oct

  Oct 8, 2026Announcements

• Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN Agents
  arxiv-ai· 09-oct

  arXiv:2610.10635v1 Announce Type: new Abstract: Aerial vision-and-language navigation (VLN) agents are typically trained on detail-rich, trajectory-aligned commands, whereas users issue short, intent-driven instructions; on a frozen OpenFly navigator, this \emph{instruction gap} drops success rate (SR) from $31.03\%$ to $11.33\%$. To scale translator training, we prompt a language model with human-written style examples to convert original commands into paired, intent-centered Weak commands, which yield $15.27\%$ SR. We introduce the \textbf{Trajectory-Grounded Instruction Translator (TGIT)}, a front-end that keeps the navigator frozen and translates Weak inputs into agent-executable commands by learning from its trajectory outcomes. The resulting Weak-trained translator raises Weak-input…

• Plan-and-Patch: Diffusion Language Models for Agentic Planning
  arxiv-ai· 09-oct

  arXiv:2610.10786v1 Announce Type: new Abstract: Planning is increasingly important for long-horizon agents, where successful execution requires coordinating subgoals, tool use, and intermediate outcomes over many steps. Yet assumptions made during planning may be invalidated by the environment, tools may return unexpected results, or actions may fail. Effective agents must therefore not only generate plans, but also revise them. Such revisions often affect only part of a plan, leaving the preceding and subsequent structure intact. Rather than regenerate the entire plan and risk unnecessary changes, repair can regenerate the affected region conditioned on the preserved prefix and suffix. We introduce Plan-and-Patch, a plan-and-act framework in which a diffusion language model (dLLM) genera…

• Whose Ground Truth? Embracing Ambiguity in Human-Centered AI
  arxiv-ai· 09-oct

  arXiv:2610.10805v1 Announce Type: new Abstract: As AI systems increasingly interact with people and make decisions about them, understanding human interpretations becomes an important part of developing human-centered AI. Conventional machine learning and AI systems are largely developed under the assumption that a single definitive ground truth exists, with variability in human annotations often resolved through aggregation or treated as noise. However, for many human-centered tasks, human interpretation is inherently ambiguous, and multiple interpretations of the same input may be simultaneously reasonable and valid. Reducing such ambiguity to a single target risks overlooking meaningful information about the diversity of human perception, judgment, and experience. In this position pape…

• On the Clock: Towards Punctual and Productive Time-Budgeted AI Agents
  arxiv-ai· 09-oct

  arXiv:2610.10833v1 Announce Type: new Abstract: We study whether small LLM agents can operate effectively under explicit wall-clock time budgets by both respecting the allocated runtime and using available time productively. We evaluate Qwen3.6-27B on five competitions from MLE-Bench Lite and Qwen3-4B on Zork I (Jericho), two agentic benchmarks where additional computational time can meaningfully improve performance. In the simplest setting, where the budget is stated only in the prompt, agents fail to translate the stated budget into controlled use of time. These failures arise from gaps in time awareness, since the harness provides no timing feedback, but also because they cannot reliably anticipate the duration of actions, and do not have a learned mapping from available time to an app…

• Self-Supervised Keyframe Discovery for Horizon-Invariant Behavior Cloning
  arxiv-ai· 09-oct

  arXiv:2610.10857v1 Announce Type: new Abstract: Behavior cloning (BC) in non-Markovian environments is a challenging problem because policies have to reason over contextual information over long horizons. Existing policy architectures rely on recurrent or attention-based mechanisms to capture long-term dependencies. However, recurrent models suffer from hidden-state collapse and gradient instability under backpropagation through time, while attention-based models are fundamentally limited by context length. To address these issues, we propose Keyframe Mnemonics, a novel self-supervised method that $\textit{discovers}$ a set of information-critical observations ($\textit{mnemonics}$) by learning an objective from randomly sampled past observations and using it as a reward for keyframe sele…

• An Explainable Header-Centric Framework for Large-Scale Semantic Table Interpretation and Data Quality Assessment
  arxiv-ai· 09-oct

  arXiv:2610.10541v1 Announce Type: new Abstract: Knowledge Graph (KG) quality depends not only on downstream graph validation, but also on the quality of tabular metadata used before integration. In metadata-only Semantic Table Interpretation (STI), where cell values are unavailable, noisy, or unsuitable, column headers become a critical source of semantic evidence for traceable KG preparation. We present an explainable, header-centric framework for metadata-only Column Type Annotation (CTA) and Data Quality Assessment (DQA). The framework maps headers to 39 interpretable FinalFormat types using curated lexical resources and preserves token-level traceability through SourceKeywords. Each assigned type activates validation rules based on a taxonomy of Data Quality Issues (DQIs), producing d…

• Synthesis Through Simulation: Generating Coherent Enterprise Data via Scalable Agent-System Interaction
  arxiv-ai· 09-oct

  arXiv:2610.10549v1 Announce Type: new Abstract: Tool-calling agents have become central to enterprise AI, yet training and evaluating them at scale remains severely constrained due to business and legal restrictions on enterprise systems, data, and database schemas. Tabular data synthesis offers a natural alternative, but its effectiveness is fundamentally limited by structural validity and schema availability, while procedure-based approaches yield the opposite weakness, typically lacking distributional fidelity without per-domain authoring. We introduce **Synthesis Through Simulation** (STS), a **schema--free** data synthesis paradigm in which an LLM agent generates data by executing operations against policy-enforcing APIs within simulated enterprise environments. Because data is gener…

• Agent-Controlled Forgetting for Tool-Using Agents: Reversible Context Curation in Practice
  arxiv-ai· 09-oct

  arXiv:2610.10590v1 Announce Type: new Abstract: Tool-using agents repeatedly carry observations whose useful content can be much smaller than their original payload. We study agent-controlled forgetting: the acting model selects previously observed tool results, replaces each with a short note at its original position, and retains the exact original in a recoverable archive. A Python harness exposes batch archival and explicit recovery without task-specific model training, while protecting user instructions and assistant messages from these operations. In an exploratory OpenTelemetry debugging case followed by an unrelated implementation task, the method ended with 231,951 provider-reported prompt tokens versus 912,492 under retained history, used 50% fewer cumulative input tokens, and ha…

• Verification and Self-Improvement in Agentic AI: Foundations and Limits
  arxiv-ai· 09-oct

  arXiv:2610.10611v1 Announce Type: new Abstract: Agentic AI systems can improve by searching longer, receiving additional support, or modifying how they propose and verify outputs. A performance score does not distinguish these mechanisms. We compare these changes through bounded verification with hidden terminal randomness. A stage specifies admissible transcripts, polynomial bounds, an alternating verification protocol, and a terminal checker. Its native reach uses default support; its closure frontier permits all support already admitted by the interface. Under a uniform pointwise probability gap and task-relative soundness, these are well-defined languages. We prove that independent majority amplification preserves both languages, whereas existential acceptance over random tapes can ad…

• The Harness as the Only Mutable Surface: Compliance-Bounded Self-Evolution of LLM Agents in Credit Pipelines, with a Measured Admission Gate
  arxiv-ai· 09-oct

  arXiv:2610.10629v1 Announce Type: new Abstract: Self-improving LLM agents can adapt a credit pipeline to a changed rule, but an agent that rewrites itself destroys the artefact a supervisor reviews: a named change, a recorded test, an approval. We argue that self-evolution is reviewable only if it is confined to the runtime harness (instruction text, tool-call logic and primitive composition) while model weights stay fixed, so that every adaptation is a diff with a cause and a test attached. We give a dual-loop engine built on that bound, with one admission gate that writes a hash-chained record before deployment, and we measure the gate in simulation, with a simulated agent and a seeded-search proposer rather than language models. Across three families of supervisory re-interpretation at…

• Why LLMs Agree With You Even When You’re Wrong
  bytebytego· 06-oct

  When LLMs sometimes agree with incorrect claims, it is mostly because their training rewards such behaviour. This reward system is built on several…Oct 6 • ByteByteGo252711

• EP228: How SSH Works
  bytebytego· 03-oct

  Secure shell is a method to access a remote machine securely over an unsecured network. It starts with a TCP connection from the SSH client to the…Oct 3 • ByteByteGo25218

• How Software Boundaries Are The Most Important Skill for Developers
  bytebytego· 08-oct

  In this article, we’ll learn some key strategies that can help a developer make the right boundary-related decisions for the systems they are working…18 hrs ago • ByteByteGo695

• How Netflix Taught an LLM to Recommend Movies So That You Keep Watching
  bytebytego· 07-oct

  In this article, we will look at how GenRec was built.Oct 7 • ByteByteGo21834

• Last 3 days: AI Evals, October cohort
  bytebytego· 04-oct

  Secure your spot in AI Evals in Practice. Enrollment closes in just 3 days! If you’ve been thinking about joining, now’s the time.Oct 4 • ByteByteGo1504

• The LLM Blindspot: Why Models Forget What’s in the Middle of Your Prompt
  bytebytego· 05-oct

  In this article, we’ll look at why LLMs have this bias against middle information.Oct 5 • ByteByteGo24679

• If your team is happy, are you doing a good job?
  charity-majors· 08-oct

  Brilliant jerks, ZIRP-era managers, and how psychological safety lost the plot. Part 3 of my conversation with Dr. Cat Hicks.

• Claude Platform release notes
  claude-changelog

  Release notesSubscribeCopy pageUpdates to the Claude Platform, including the Claude API, client SDKs, and the Claude Console.SubscribeCopy pageThe Claude Platform release notes list changes to the Claude API, the client SDKs, and the Claude Console, newest first. For release notes on Claude Apps, see the Release notes for Claude Apps in the Claude Help Center.For updates to Claude Code, see the complete CHANGELOG.md in the claude-code repository. October 8, 2026 The Compliance API chat endpoints now also return chats from the unified Claude experience, in beta for Claude Enterprise organizations, with your existing Compliance Access Key. See Retrieve and delete chats, files, and projects. October 7, 2026 We've lowered the price of prompt cache reads on Claude Sonnet 5.5 from $0.20 …

• Bridging technical depth and usability: The story behind Radar’s redesign
  cloudflare· 08-oct

  We redesigned Cloudflare Radar to make real-time global traffic and outage data accessible to a wider audience — including journalists, less technical researchers, and everyday users. By introducing an interactive map and standardized components, we simplified navigation while preserving technical depth.

• The keys to the Internet change on October 11. Are you ready?
  cloudflare· 06-oct

  On October 11, 2026, the DNS root switches to a new key-signing key (KSK-2024). Learn what this means for you, and how RFC 8509 trust anchor sentinels allow you to test whether your DNS resolver is ready for the rollover.

• Streamline: custom video pipelines with Cloudflare Stream and Workers
  cloudflare· 02-oct

  Streamline demonstrates how to build long-running, continuous video processing pipelines by pairing Cloudflare Workers and Durable Objects with a containerized media engine.

• 8 major updates to Cloudflare Observability
  cloudflare· 02-oct

  Cloudflare is launching eight major updates that bring logs, traces, analytics, alerts, dashboards, querying, and telemetry export into one observability platform, with simpler and more predictable pricing.

• One year later: the power of 1.1.1.1 interns
  cloudflare· 05-oct

  A year after announcing our goal to hire 1,111 interns, more than 750 early-career builders have shipped real products across 48 teams at Cloudflare. From Birthday Week launches to post-quantum security, our interns prove that AI amplifies human potential instead of replacing it.

• Building an evidence-grounded agentic security operations harness on Cloudflare
  cloudflare· 07-oct

  Cloudflare Managed Defense uses a team of specialized AI agents built on Workers and global network telemetry to analyze security alerts. By separating deterministic evidence collection from model inference, the system delivers grounded recommendations to Managed Defense Analysts.

• Everything we launched during Birthday Week 2026
  cloudflare· 05-oct

  We celebrated our 16th birthday with 46 announcements across open source, post-quantum security, AI agents, and developer platform upgrades. Here’s a day-by-day roundup of everything we shipped.

• Introducing Cloudflare Traces: follow requests through our entire platform
  cloudflare· 02-oct

  Cloudflare Traces shows how a request moves through security rules, transformations, cache, routing, Workers, and your origin, then follows it across services running anywhere in your stack.

• Updates on our pledge to make Cloudflare features accessible to everyone
  cloudflare· 02-oct

  A year after pledging to eliminate two-tier product access, Cloudflare has expanded Logpush, multi-account governance, and higher platform limits to all accounts. Here is an update on our progress, how we dogfood these tools internally, and what is coming next.

• Introducing Web Search API via AI Gateway
  cloudflare· 02-oct

  Cloudflare AI Gateway now supports native web search API integration in partnership with Ceramic.ai, Exa, and Linkup. Developers can now inject real-time web context into model inference calls via AI Gateway, REST APIs, or Workers bindings.

• cathrynlavery/diagram-design
  github-trending

  Editorial diagram design for Claude Code, Codex, GitHub Copilot, Factory Droid, and Pi. 42 diagram types. Self-contained HTML + SVG. No shadows. No Mermaid slop. https://cathrynlavery.github.io/diagram-design/ New in 2.0 — the Loop: flywheels with a shared-memory hub. The dashed lines are the write-backs. New in 2.3: semantic system patterns and optional accessible motion, while static output stays the default. New in 2.5.10: ten more layout grammars — Sankey, fishbone, Wardley map, kanban, user journey, deployment, dependency graph, UML class, story map, and database schema. Editorial diagram types for Claude Code, Codex, Factory Droid, Pi, and Agent Skills-compatible hosts. Self-contained HTML + SVG. No shadows. No Mermaid slop. Semantic patterns describe behavior separately from layout…

• thedotmack/claude-mem
  github-trending

  Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More https://claude-mem.ai 🇨🇳 中文 • 🇹🇼 繁體中文 • 🇯🇵 日本語 • 🇵🇹 Português • 🇧🇷 Português • 🇰🇷 한국어 • 🇪🇸 Español • 🇩🇪 Deutsch • 🇫🇷 Français • 🇮🇱 עברית • 🇸🇦 العربية • 🇷🇺 Русский • 🇵🇱 Polski • 🇨🇿 Čeština • 🇳🇱 Nederlands • 🇹🇷 Türkçe • 🇺🇦 Українська • 🇻🇳 Tiếng Việt • 🇵🇭 Tagalog • 🇮🇩 Indonesia • 🇹🇭 ไทย • 🇮🇳 हिन्दी • 🇧🇩 বাংলা • 🇵🇰 اردو • 🇷🇴 Română • 🇸🇪 Svenska • 🇮🇹 Italiano • 🇬🇷 Ελληνικά • 🇭🇺 Magyar • 🇫🇮 Suomi • 🇩🇰 Dansk • 🇳🇴 Norsk Persistent memory compression system built for Claud…

• EpicGames/raddebugger
  github-trending

  A native, user-mode, multi-process, graphical debugger. The RAD Debugger Project NOTE: This README does not document usage instructions and tips for the debugger itself, and is intended as a technical overview of the project. The debugger's README, which includes usage instructions and tips, can be found packaged along with debugger releases, or within the build folder after a local copy has been built. You can find pre-built release binaries here. The RAD Debugger is a native, user-mode, multi-process, graphical debugger. It currently only supports local-machine Windows x64 debugging with PDBs, with plans to expand and port in the future. In the future we'll expand to also support native Linux debugging and DWARF debug info. The debugger is currently in ALPHA. In order to get the debugge…

• liquidslr/system-design-notes
  github-trending

  Notes of the book System Desgin Interview - An Insider's Guide https://pagefy-blush.vercel.app/system-design/system-design-interview-by-alex-xu System Design Interview - An Insider's Guide (Vol 1 and 2) These notes are based on the System Design Interview books - Vol 1 and Vol 2 2nd Ed Check the notes here: https://pagefy.io/system-design/system-design-interview-by-alex-xu Note: These notes are a work in progress. Chapter 1 - Scale From Zero To Millions Of Users Chapter 2 - Back-of-the-envelope Estimation Chapter 3 - A Framework For System Design Interviews Chapter 4 - Design A Rate Limiter Chapter 5 - Design Consistent Hashing Chapter 6 - Design A Key-Value Store Chapter 7 - Design A Unique ID Generator In Distributed Systems Chapter 8 - Design A URL Shortener Chapter 9 - Design A Web Cr…

• morluto/rea
  github-trending

  Reverse engineer anything with agents, from app behavior down to native binaries. https://rea.tools English · 简体中文 · 繁體中文 · 日本語 · 한국어 · Türkçe · Русский · Tiếng Việt · ไทย · Deutsch · Español · Français · Українська · Polski · Português (Brasil) · العربية · فارسی REA: Reverse Engineer Anything One MCP for reverse engineering across binaries, applications, and runtime behavior. See a feature you like. Understand how it works, down to the binary level. Website · Guides · Showcases Quick start · How REA works · What you can analyze · Showcases · FAQ · Documentation npx rea-agents setup Join the Reverse Engineering Community Discord · Q&A · Show and Tell See a feature in an app that you want in your own product? Ask your agent to investigate it with REA. It can inspect the app without its sou…

• storytold/artcraft
  github-trending

  ArtCraft is an intentional crafting engine for artists, designers, and filmmakers https://getartcraft.com ArtCraft The IDE for artists. https://github.com/user-attachments/assets/7339bd09-3e0d-44a7-a024-af9de64e4c11 Watch the two-minute ArtCraft demo on YouTube Download ArtCraft · GitHub releases · Build from source ArtCraft is the IDE for interactive AI image and video creation. Compose in 2D, stage scenes in 3D, and choose the models that fit your work. Turn prompting into crafting with visual tools for precise, repeatable results. Show, Don't Tell: Advanced Crafting Features Artists need control over composition, character identity, and camera placement. Build the scene before you generate it. Click any preview to view it at full size. Feature Preview Image to Location Place virtual ac…

• anthropics/knowledge-work-plugins
  github-trending

  Open source repository of plugins primarily intended for knowledge workers to use in Claude Cowork Knowledge Work Plugins Plugins that turn Claude into a specialist for your role, team, and company. Built for Claude Cowork, also compatible with Claude Code. Why Plugins Cowork lets you set the goal and Claude delivers finished, professional work. Plugins let you go further: tell Claude how you like work done, which tools and data to pull from, how to handle critical workflows, and what slash commands to expose — so your team gets better and more consistent outcomes. Each plugin bundles the skills, connectors, slash commands, and sub-agents for a specific job function. Out of the box, they give Claude a strong starting point for helping anyone in that role. The real power comes when you cus…

• boykopovar/AnyPS5
  github-trending

  Tool for automatic PS5 executables porting to Linux and Windows https://discord.gg/BHFztBPUe About Tool for automatic executables porting to Linux and Windows. Includes a relinker that converts executable to the target system's native format and implementations of system prx libraries suitable for dynamic linking. No emulation or separate runtime process. Usage, Build instructions, Architecture, Technical debt of the project, code style conventions, contributing Status * System libraries: percentage of the functions known to the project so far (declared in core/libs/prx), not of every PS5 system function. The total grows as more functions are declared. List of verified games Dreaming Sarah (2D platformer) runs at a stable 60 fps on a GTX 1050 Ti / i5-7500 3.4GHz. Unsupported or unexpected…

• The looming AI consciousness trainwreck (2022)
  hn-ai· 09-oct

  Article URL: https://blog.plan99.net/the-looming-ai-consciousness-train-wreck-2e23165b3246 Comments URL: https://news.ycombinator.com/item?id=50018146 Points: 1 # Comments: 0

• TokenRouter: A serving engine for token-level LLM routing
  hn-ai· 09-oct

  Article URL: https://fuvty.github.io/thinking_yard_project_page/projects/tokenrouter/ Comments URL: https://news.ycombinator.com/item?id=50017997 Points: 1 # Comments: 0

• If your team is too busy doing their 'job' to experiment with AI
  hn-ai· 09-oct

  Article URL: https://ghuntley.com/replaced/ Comments URL: https://news.ycombinator.com/item?id=50018057 Points: 1 # Comments: 0

• codyssey: turn every Claude Code session into a little adventure
  hn-ai· 09-oct

  Article URL: https://github.com/delexw/codyssey Comments URL: https://news.ycombinator.com/item?id=50017890 Points: 1 # Comments: 0

• Show HN: Long term Memory and 50M token window for LLM
  hn-ai· 09-oct

  Article URL: https://huggingface.co/papers/2610.10845 Comments URL: https://news.ycombinator.com/item?id=50018178 Points: 2 # Comments: 0

• Kubernetes for AI agents: when code is cheap, proof matters
  hn-ai· 09-oct

  Article URL: https://kubedex.com/kubernetes-ai-agentic-development/ Comments URL: https://news.ycombinator.com/item?id=50017698 Points: 1 # Comments: 0

• Show HN: VendIQ – AI agent for vendor follow-ups and ops automation
  hn-ai· 09-oct

  Article URL: https://vendiq.rishailabs.com/ Comments URL: https://news.ycombinator.com/item?id=50018161 Points: 1 # Comments: 0

• Aleph Alpha and Mistral launches fuel Europe's sovereign AI push
  hn-ai· 09-oct

  Article URL: https://www.renascence.io/news/96889/aleph-alpha-mistral-launches-fuel-europes-sovereign-ai-push Comments URL: https://news.ycombinator.com/item?id=50017858 Points: 2 # Comments: 1

• Teen hike Claude Canada wilderness - emergency rescue
  hn-ai· 09-oct

  Article URL: https://www.theguardian.com/world/2026/oct/08/teen-hike-claude-ai-directions-rescue-canada-mountain Comments URL: https://news.ycombinator.com/item?id=50017771 Points: 2 # Comments: 0

• [video] Pedagogy in the Times of AI
  hn-ai· 09-oct

  Article URL: https://www.youtube.com/watch?v=N2a1J0UPeL4 Comments URL: https://news.ycombinator.com/item?id=50017843 Points: 1 # Comments: 0

• [AINews] not much happened today
  latentspace· 08-oct

  a quiet day.

• [AINews] Reflection Beam - 501B-A23B American Open Model
  latentspace· 06-oct

  A small win for US open source

• Inside-Out AI: Rebuilding Airbnb Behind the Scenes and Across the Guest Experience
  latentspace· 02-oct

  After leading Meta’s Llama models, Ahmad Al-Dahle is now transforming Airbnb with AI — from how its teams develop products to how it serves guests.

• Synthesis Superintelligence: from Semiconductors to Superconductors — Periodic Labs’ Liam Fedus and Ekin Dogus Cubuk
  latentspace· 08-oct

  A special Science pod and Engineering pod crossover.. with Forward Deployed Engineering kicker!

• Can a Cloud-Native Harness Make Agents Reliable Beyond the Desktop?
  latentspace· 07-oct

  Kubernetes co-creators Craig McLuckie and Joe Beda aim to bring agent harnesses fully into the cloud.

• [AINews] Quasi-Riemann-Hypothesis: OpenAI publishes 722 math papers solving 90 of the top 500 open math problems; “the most significant moment” in >100 years of mathematics
  latentspace· 07-oct

  Our head hurts.

• [AINews] Claude Haiku 5.5 — better than GPT-6 Luna at the same pricing
  latentspace· 08-oct

  yay small models

• Fragments: October 4
  martin-fowler· 05-oct

  In response to my last fragments (probably the bit about us worrying if LLMs have consciousness when we when we should be wondering why they don’t have a conscience) “Metalanguage” replied: we shipped the id and forgot the superego. classic software lifecycle. I don’t know what was on their mind, but their post immediately made me think of the classic 1956 movie Forbidden Planet. Plenty of sci-fi, and other literature, have explored humans creating technology with unintended behavior, going back at least to Mary Shelly. But that movie was particularly influential on sci-fi film-making and in the heart of its story is what happens when we nurture a thinking machine. I use the term “nurture” here deliberately. We talk of building software, but building implies a degree of determinism. When …

• OpenAI “rogue” agent activities found on Wikimedia projects
  simonw· 07-oct

  <p><strong><a href="https://wikimediafoundation.org/news/2026/10/05/openai-rogue-agent-activities-found-on-wikimedia-projects/">OpenAI “rogue” agent activities found on Wikimedia projects</a></strong></p> Given how tempting a target wikis are for rogue agent swarms, it's not a huge surprise that Wikipedia found evidence of that activity once they went looking:</p> <blockquote> <p>The Wikimedia Foundation conducted its own investigation to see whether Wikimedia websites had been similarly affected by AI agents, focusing on those operated by OpenAI. We can confirm that we have discovered some activity by these “rogue” OpenAI agents on Wikimedia platforms. The unauthorized bot activities included edits to our wikis, some unsuccessful attempts to exploit a public note-taking tool we host, and…

• ttok 0.4
  simonw· 08-oct

  <p><strong>Release:</strong> <a href="https://github.com/simonw/ttok/releases/tag/0.4">ttok 0.4</a></p> <p><code>ttok</code> is my CLI tool for counting tokens, using OpenAI's open source <a href="https://github.com/openai/tiktoken">tiktoken</a> library.</p> <p>It hasn't been in updated in a couple of years, but I finally fixed a Click warning, updated CI, and added a <code>--list-models</code> command to list available models.</p> <p>It works with <code>uvx</code>, so you can count tokens in anything like this:</p> <pre><code>cat file.txt | uvx ttok </code></pre> <p>Tags: <a href="https://simonwillison.net/tags/projects">projects</a>, <a href="https://simonwillison.net/tags/ai">ai</a>, <a href="https://simonwillison.net/tags/openai">openai</a>, <a href="https://simonwillison.net/tags/gen…

• Quoting Carson Gross
  simonw· 08-oct

  <blockquote cite="https://htmx.org/essays/yes-and/"><p>Computer programming is, fundamentally, about two things:</p> <ul> <li>Problem-solving using computers</li> <li>Learning to control complexity while solving these problems</li> </ul> <p>I have a hard time imagining a future where knowing how to solve problems with computers and how to control the complexity of those solutions is <em>less</em> valuable than it is today, so I think it will continue to be a viable career even with the advent of AI tools.</p></blockquote> <p class="cite">&mdash; <a href="https://htmx.org/essays/yes-and/">Carson Gross</a></p> <p>Tags: <a href="https://simonwillison.net/tags/computer-science">computer-science</a>, <a href="https://simonwillison.net/tags/carson-gross">carson-gross</a>, <a href="https://simon…

• ttok 1.0
  simonw· 09-oct

  <p><strong>Release:</strong> <a href="https://github.com/simonw/ttok/releases/tag/1.0">ttok 1.0</a></p> <p>I released <a href="https://simonwillison.net/2026/Oct/8/ttok/">ttok 0.4</a>, ran <code>uv tool upgrade ttok</code>, piped a file into the new version... and realized that it was defaulting to the GPT-4 tokenizer when it should very clearly default to GPT-5/GPT-6 instead!</p> <p>I figured switching the default was a reasonable excuse to finally ship a 1.0.</p> <p>OpenAI haven't actually confirmed that GPT-6 uses the same tokenizer as the GPT-5 family yet - there's an <a href="https://github.com/openai/tiktoken/issues/608">angry issue about it</a> - but I found <a href="https://github.com/williamliu-ai/token-count-compare/commit/8d8a2178538bb37beb548fc37378135e4ff4b0c8">this commit</a…

• Quoting Ben Affleck
  simonw· 07-oct

  <blockquote cite="https://www.youtube.com/watch?v=HzraLR2VkNM&amp;t=2259s"><p>I've always been kind of into computers since I was young. And then when film started to move from analog film to digital, I became more interested in that aspect of it. And the visual effects workflow for many years has included machine learning.</p> <p>So I can write like pretty shitty Python scripts and stuff like that because with convolutional neural networks, which were the sort of precursors to what the transformer can do, which is just much more computation simultaneously, you would do things like look at what's called a tensor, which is just the numerical translation of a visual image in numbers — like the batch number, the frame number, the red, green, and blue values of each pixel in each frame. </p> …

• Anti-Patterns in Software Blogging
  simonw· 07-oct

  <p><strong><a href="https://refactoringenglish.com/blog/anti-patterns-software-blogging/#summary">Anti-Patterns in Software Blogging</a></strong></p> Some excellent writing advice from Michael Lynch. Michael warns against "meandering intros", misjudging your reader's existing knowledge, assuming they'll read your previous posts, and excessive formality.</p> <p>He also warns against <a href="https://refactoringenglish.com/blog/anti-patterns-software-blogging/#overreliance-on-links">overreliance on links</a> as an excuse not to explain terminology. This one hurt! I do this all the time, but I have a nagging suspicion that almost nobody ever clicks on them.</p> <p>(In <a href="https://lobste.rs/s/rwdufq/anti_patterns_software_blogging#c_qdhhsd">a Lobste.rs comment</a> Michael clarifies that …

• Quoting Victoria Kim
  simonw· 06-oct

  <blockquote cite="https://www.nytimes.com/live/2026/10/05/world/openai-australia-hearing/94c8067e-4098-5c00-9673-c78cb58a9d8a"><p>Since the Medicare breach, OpenAI has put in place additional monitoring to allow “immediate intervention” by staff to stop training if the company’s models access the internet in ways they’re not supposed to, Mr. Kwon [chief strategy officer at OpenAI] said.</p></blockquote> <p class="cite">&mdash; <a href="https://www.nytimes.com/live/2026/10/05/world/openai-australia-hearing/94c8067e-4098-5c00-9673-c78cb58a9d8a">Victoria Kim</a>, Reporting from the Australian parliament</p> <p>Tags: <a href="https://simonwillison.net/tags/accidental-cyberattacks">accidental-cyberattacks</a>, <a href="https://simonwillison.net/tags/generative-ai">generative-ai</a>, <a href="h…

• llm-openai-decisions 0.1a0
  simonw· 06-oct

  <p><strong>Release:</strong> <a href="https://github.com/simonw/llm-openai-decisions/releases/tag/0.1a0">llm-openai-decisions 0.1a0</a></p> <p>OpenAI released their new Jev-style <a href="https://developers.openai.com/api/docs/guides/decisions">Decisions API</a>, as previously announced at last week's DevDay.</p> <p>Since I already have an <a href="https://github.com/simonw/llm-typesafe">llm-typesafe</a> plugin for talking to Jev, I had GPT-6 Astra read the new OpenAI API documentation and build an <code>llm-openai-decisions</code> plugin inspired by <code>llm-typesafe</code>.</p> <p>Unlike Jev, the new <code>gpt-6-luna</code> decision model supports image input in addition to text. Both models charge for input it and not for output: OpenAI's is 10 cents per million input tokens, Jev's is…

• Claude Haiku 5.5
  simonw· 07-oct

  <p>As previously <a href="https://simonwillison.net/2026/Sep/28/claude-sonnet-5-5/">promised</a>, here's Anthropic's new fast, low cost model: <a href="https://www.anthropic.com/claude-haiku-5-5">Introducing Claude Haiku 5.5</a>.</p> <p>The previous Haiku, 4.5, was very much showing its age. It came out <a href="https://simonwillison.net/2025/Oct/15/claude-haiku-45/">almost a year ago</a>, and was priced at $1/million input and $5/million output - relatively expensive even back then, and a full 10x the price of OpenAI's <a href="https://simonwillison.net/2026/Sep/22/opus-and-sol-and-luna/#gpt-6-sol-and-luna-are-half-the-price-of-their-gpt-5-6-equivalents">GPT-6 Luna</a>, released last month.</p> <p>The new Haiku exactly matches the price of GPT-6 Luna - $0.10/$0.50 - up to 100,000 tokens.…

• Quoting Jake Boggan
  simonw· 07-oct

  <blockquote cite="https://news.ycombinator.com/item?id=49984923#49987367"><p>I was a graph theory junkie long ago and even moved to Budapest for awhile to study among the greats. While I was there I started working on Barnette's Conjecture which came to occupy my thoughts over the next 24 years of my life, on and off as I worked in many different fields. Last summer I even thought for a few days that I had actually solved it.</p> <p>But it's supposedly proven here - <a href="https://github.com/openai/math/blob/main/lean/docs/180.md">problem 180</a>. I don't know what to think exactly. I spent thousands of hours on that problem. I really enjoyed it. Hearing that it is solved somehow makes me sad in a far-off way, like hearing an ex-girlfriend died suddenly in a car crash. I don't know, the…