GenAI PMSimon Willison
A developer and AI commentator quoted here in relation to OpenAI’s clarification of ChatGPT Work behavior. He is relevant as an interpreter and critic of product messaging.
Key Highlights
- Simon Willison is a high-signal interpreter of AI model launches, developer tools, and product messaging for technical audiences.
- He is especially relevant to AI PMs because he tests claims in practice and explains what launches mean for real workflows.
- His coverage of prompt injection and role confusion provides tactical insight into security and trust risks in LLM products.
- He often surfaces the product implications of local models, coding agents, APIs, and open-weight releases before consensus forms.
- His commentary on ChatGPT Work illustrates his value as a critic of ambiguous vendor explanations about execution and data boundaries.
Simon Willison
Overview
Simon Willison is a developer, writer, open-source builder, and one of the most consistently useful interpreters of the fast-moving AI product landscape. In this corpus, he appears both as a primary source of analysis on model releases, tooling, security issues, and developer workflows, and as a quoted commentator on product messaging—especially where vendors' claims about AI behavior, deployment boundaries, or capabilities need clearer scrutiny.For AI Product Managers, Simon Willison matters because he sits at the intersection of hands-on experimentation and public explanation. He does not just repeat launch claims; he tests products, surfaces edge cases, links technical details to real usage, and often frames what a release actually means for builders. That makes him a valuable signal for PMs tracking model quality, agentic coding workflows, prompt injection risks, local-vs-cloud execution tradeoffs, and the gap between marketing language and product reality.
Key Developments
- 2026-06-16 — Also cited in coverage of Anthropic’s Fable 5 / Mythos 5 rollback after a public jailbreak, indicating his relevance in fast-moving safety and policy discussions around model availability.
- 2026-06-17 — Highlighted a Hacker News comment from Georgi Gerganov praising Qwen3.6-27B as a practical local coding model, reinforcing Simon’s role in surfacing credible practitioner feedback on local LLM workflows.
- 2026-06-18 — Published commentary that GLM-5.2 was likely the most powerful text-only open-weights LLM at the time, emphasizing its scale, open weights, MIT license, and long context window.
- 2026-06-23 — Covered Prompt Injection as Role Confusion, explaining research that models can mistake privileged instructions for user content based on writing style, and noting mitigation via “destyling.” This is a high-signal security interpretation for AI builders.
- 2026-06-29 — Amplified Jon Udell’s “Agent in the loop” framing, arguing against vague “human in the loop” language and toward agent collaboration embedded in existing workflows.
- 2026-06-30 — Wrote about Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding, summarizing model variants, coding performance, and successful local runs via LM Studio and Pi.
- 2026-07-07 — Covered Tencent’s open-source Hy3 Mixture-of-Experts release, noting its Apache 2.0 license, FP8 quantization, long context, and trial access via OpenRouter.
- 2026-07-08 — Shared an experimental github-code Web Component built with GPT-5.5 for embedding GitHub code snippets in the web, demonstrating his practical interest in lightweight developer UX patterns around AI-assisted tooling.
- 2026-07-10 — Reported on Meta’s Muse Spark 1.1, focusing on API availability and claimed improvements in agentic tool-calling and computer use.
- 2026-07-11 — Quoted OpenAI attempting to clarify ChatGPT Work behavior: web and mobile run in the cloud, while desktop can use local files and remain on-device; he was cited as interpreting this clarification as still unsuccessful, underscoring his role as a critic of ambiguous product messaging.
Relevance to AI PMs
1. Use him as an early-warning interpreter of launches and claims. Simon’s coverage is often most useful after a flashy release announcement, when PMs need to know what actually changed: API access, licensing, context length, local usability, tool use, or deployment constraints.2. Track his writing for practical security and reliability implications. His attention to topics like prompt injection, role confusion, jailbreaks, and unclear execution boundaries helps PMs anticipate user trust issues, enterprise objections, and product requirement gaps before they become incidents.
3. Apply his workflow lens to roadmap decisions. Simon frequently focuses on real developer usage: local models, agentic coding, browser-based experimentation, and composable tooling. PMs can use that perspective to prioritize features that fit existing workflows instead of forcing fully autonomous agent experiences users may not trust.
Related
- OpenAI — Connected through commentary on ChatGPT Work behavior and interpretation of product clarification.
- ChatGPT Work — A key example of Simon’s relevance as a critic of vague messaging around cloud vs. local execution.
- LLM / llm-python-library — Strongly associated with Simon’s broader role in the developer LLM ecosystem and practical experimentation.
- Prompt Injection — A major theme in his coverage, especially relevant for product safety, trust, and enterprise readiness.
- Agentic coding / coding agents / agentic engineering — He is frequently cited around how coding models and agents perform in real workflows.
- Anthropic / Claude / Claude Code — Related through recurring coverage of model behavior, coding workflows, and safety incidents.
- Qwen, Gemma, GLM-5.2, Hy3, Ornith-1.0, Muse Spark 1.1 — Examples of model releases where Simon functions as a translator between raw technical announcement and practical product significance.
- Jon Udell, Georgi Gerganov, Hacker News — Reflect his role in curating credible technical discourse and elevating operator-level insights for a wider audience.
Newsletter Mentions (83)
“OpenAI - Quote from OpenAI attempting to clarify ChatGPT Work behavior: web and mobile Work run in the cloud, desktop Work can use local files and remains on the desktop; author notes the clarification was unsuccessful.”
#13 📝 Simon Willison OpenAI - Quote from OpenAI attempting to clarify ChatGPT Work behavior: web and mobile Work run in the cloud, desktop Work can use local files and remains on the desktop; author notes the clarification was unsuccessful. #14 𝕏 Rowan Cheung : Meta dropped Muse Spark 1.1, an agentic AI that outperforms OpenAI and Anthropic while massively undercutting their prices—proof that Zuckerberg’s tens-of-billions-dollar compute bet is paying off. Meta’s stock has jumped over 10% since the release.
“Simon Willison Introducing Muse Spark 1.1 - Meta released Muse Spark 1.1, the first Spark model to offer an API, with claimed improvements in agentic tool calling and computer use.”
Simon Willison is credited with multiple newsletter items, including model analyses and release commentary.
“Simon Willison github-code Web Component - An experimental Web Component was built (using GPT-5.5) to embed code from GitHub URLs by converting them to raw.githubusercontent links and fetching/displaying specified ranges of lines with line numbers.”
#4 📝 Simon Willison github-code Web Component - An experimental Web Component was built (using GPT-5.5) to embed code from GitHub URLs by converting them to raw.githubusercontent links and fetching/displaying specified ranges of lines with line numbers.
“Tencent releases open-source Hy3 MoE model with FP8 quantization #1 📝 Simon Willison tencent/Hy3 - Announcement and notes about Tencent's new Hy3 Mixture-of-Experts model (Apache 2.0).”
GenAI PM Daily July 07, 2026 GenAI PM Daily 🎧 Listen to this brief 3 min listen Today's top 20 insights for PM Builders, ranked by relevance from Blogs, X, YouTube, and LinkedIn. Tencent releases open-source Hy3 MoE model with FP8 quantization #1 📝 Simon Willison tencent/Hy3 - Announcement and notes about Tencent's new Hy3 Mixture-of-Experts model (Apache 2.0). The post describes model size, availability (including a quantized FP8 variant), huge context length, a free OpenRouter trial, and includes an example SVG the author generated.
“#6 📝 Simon Willison Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding - Notes an interesting open-weights model release from DeepReinforce called Ornith-1.0 (variants include 9B Dense, 31B Dense, 35B MoE, and 397B MoE) built on Gemma 4 and Qwen 3.5, reporting strong coding performance and successful runs locally using LM Studio and Pi.”
#6 📝 Simon Willison Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding - Notes an interesting open-weights model release from DeepReinforce called Ornith-1.0 (variants include 9B Dense, 31B Dense, 35B MoE, and 397B MoE) built on Gemma 4 and Qwen 3.5, reporting strong coding performance and successful runs locally using LM Studio and Pi. Includes links to a terminal session demonstrating code-finding tasks and a pelican-drawing gist showing generation speed.
“#11 📝 Simon Willison Agent in the loop - Simon highlights a Jon Udell piece arguing we should stop saying "human in the loop" and instead view agents as teammates we invite into our established workflows, avoiding opaque agent-only processes.”
Simon Willison is the author/commentator in a post about how to frame human-agent collaboration.
“Simon Willison Prompt Injection as Role Confusion - Discussion of a paper showing models confuse privileged role-tagged text with user text based on style, enabling serious jailbreaks; the authors demonstrate that 'destyling' text can dramatically reduce attack success.”
Simon Willison is cited twice, once on prompt injection and once on porting a model to the browser with Claude Code.
“GLM-5.2 is probably the most powerful text-only open weights LLM - Chinese AI lab Z.ai released GLM-5.2 and then published the full open weights under an MIT license; it’s a 753B-parameter, Mixture-of-Experts model with a 1 million token context window and text-only inputs.”
#22 📝 Simon Willison GLM-5.2 is probably the most powerful text-only open weights LLM - Chinese AI lab Z.ai released GLM-5.2 and then published the full open weights under an MIT license; it’s a 753B-parameter, Mixture-of-Experts model with a 1 million token context window and text-only inputs.
“#19 📝 Simon Willison Georgi Gerganov - A quoted Hacker News comment from Georgi Gerganov endorses Qwen3.6-27B as a capable local model for coding tasks, describing his daily use and a lightweight harness to align it with his style.”
#19 📝 Simon Willison Georgi Gerganov - A quoted Hacker News comment from Georgi Gerganov endorses Qwen3.6-27B as a capable local model for coding tasks, describing his daily use and a lightweight harness to align it with his style.
“Also covered by: @Simon Willison”
#4 ▶️ One man just liberated Fable... and now it’s illegal Fireship The US Department of Commerce banned Anthropic’s safety-enhanced model Fable 5 (and its full-capacity sibling Mythos 5) three days after Fable’s release, forcing Anthropic to pull both models and revert users to Opus 4.8 due to a public jailbreak. Also covered by: @Simon Willison
Related
Anthropic’s coding product/blog referenced in a customer story about Cognition’s use of Claude Fable 5. For AI PMs, it highlights enterprise coding adoption narratives.
Anthropic is the company behind Claude and Claude Code. The newsletter covers its new Reflection dashboard and an enterprise deployment of Claude in industrial workflows.
OpenAI is the company behind GPT models and ChatGPT, and it appears here as the launcher of GPT-5.6 Luna and the relauncher of its Bio Bug Bounty. For AI PMs, it signals continued productization of frontier models and safety programs.
Anthropic’s assistant and coding tool, discussed here in both the Reflection dashboard and a physical-AI deployment at UST. The newsletter highlights its usage analytics, workflow suggestions, and enterprise integration.
A code editor and AI agent workspace that introduced Side Chats and cloud agent hooks in this newsletter. For AI PMs, it shows how copilots are evolving into persistent, context-aware agent threads.
A PM/influencer who shares practical AI workflow experiments around planning, design, and execution. He is cited using Fable, Claude Design, and GPT-5.6 together in a product-building workflow.
LlamaIndex is referenced as a company/brand running ParseBench against GPT-5.6. The note highlights its use in evaluating document parsing performance.
A ChatGPT-related coding/product mode discussed as a voice-and-tone setting rather than a separate product. For PMs, it highlights how users mentally bucket product experiences.
AI developer advocate and AI product communicator associated with Google DeepMind. He is credited here for announcing new Gemini API Managed Agent features.
Writer and newsletter author known for product and career analysis. He is cited here for a 2026 workforce survey about AI’s impact on sentiment.
The AI platform whose profiles are mentioned as a future personalization signal for HuggingNews. For PMs, it indicates ecosystem-based personalization and developer identity integration.
Google’s AI research lab, mentioned here in connection with interpretability and model reasoning. For PMs, it represents frontier research into understanding and auditing model behavior.
Google AI product leader frequently associated with AI Studio and developer-facing launches. Here he is credited with rolling out GitHub import in AI Studio Build.
An AI assistant or agent instance used in a public prompt-injection challenge and later in startup support automation. It is relevant to AI PMs as an example of both security testing and customer support automation.
An AI educator and researcher cited here for model-usage advice on agentic coding. He is relevant to PMs as a source of practical guidance on model selection and cost/performance tradeoffs.
Google’s AI assistant/model family, referenced here through Josh Woodward’s community feedback post. The newsletter suggests product improvements are being informed by large-scale user replies.
A product leader and commentator cited in the newsletter multiple times. She appears in the Gusto shipping story and in discussion of AI-first product development.
Technology company named as a challenger in the predicted AI super app market. It is a major platform owner and AI competitor for PMs.
CEO/founder associated with Perplexity who comments on production AI systems and model economics. He is quoted on multi-model harnesses and local model deployment timelines.
Well-known AI researcher and builder, mentioned here as joining Anthropic to use Claude for research acceleration. Relevant to AI PMs as a signal of AI-powered research workflows and talent movement.
CEO of OpenAI and a frequent commentator on model capability, economic impact, and product direction. In this newsletter he is quoted on GPT-5.6 medical reliability and AI’s net job creation so far.
Qwen is an AI model family / brand associated with open-source releases and agent infrastructure work. In this newsletter it is the source of Qwen-AgentWorld-35B-A3B and AgentWorldBench.
Meta is cited here as the source of Muse Spark 1.1 and Coding Agents guidance, emphasizing aggressive AI product and infrastructure investment. For PMs, it underscores competition on cost and capability.
Co-founder and CEO of Google DeepMind, cited unveiling DiffusionGemma. His mention ties Google’s research leadership to model launches.
Google AI leader and prominent engineering executive. Here he is cited highlighting a TPU supercomputing paper and hardware progression.
CEO of Google and Alphabet, mentioned here in connection with Gemini/DiffusionGemma announcements and open-sourcing model weights.
An OpenAI model used in the background by GPT-Live for deeper searches or reasoning. It is also mentioned as part of a multimodel harness workflow.
Systems that use models plus tools, memory, and planning to perform multi-step tasks autonomously or semi-autonomously. The newsletter references both agent architectures and agentic coding/workflows.
OpenAI’s coding agent used for autonomous implementation, browser scraping, and prototype generation in this newsletter. It is relevant for agentic coding workflows and PM-led prototyping.
A Google model described as best-in-class across hardware tiers and suitable for local on-device intelligence.
A parsing tool used to convert file and directory contents into clean, structured Markdown. It is referenced as part of an agent framework template.
A major software and cloud company referenced in relation to AI market concentration concerns. It appears as a comparator in Clem’s quote.
An AI development pattern where models act more like autonomous coding agents. The newsletter uses it to describe both NVIDIA Dynamo’s target workload and GPT-5.5/Codex improvements.
Anthropic Labs is mentioned as the organization where Henry Shi works with the founders. It appears as part of the credibility framing for the sponsored AI PM certification.
A model used as the underlying engine for an assistant tested against prompt injection. The newsletter notes its explicit anti-prompt-injection rules as a sign that defense measures are improving.
A Claude model version referenced as part of a prompt-comparison analysis. It serves as one endpoint for examining changes in Anthropic’s system prompt evolution.
Claude Opus 4.7 is a Claude model referenced for strong resistance to prompt injection in Anthropic's safety discussion. The newsletter gives specific success-rate estimates under attack attempts.
A GPT model variant used here for scientific reasoning and agentic chemistry experimentation. The newsletter frames it as a model capable of proposing experimental improvements and driving benchmarked workflows.
A Claude model used by Cognition for overnight work and production workflows. For AI PMs, it signals trust, reliability, and enterprise readiness for coding tasks.
Technology company named as a challenger in the predicted AI super app landscape. It is relevant as a potential platform competitor and distribution powerhouse.
An attack technique where malicious instructions manipulate model behavior, often by hiding within content or privileged roles. The newsletter frames role confusion as a core challenge for defending against it.
A model-routing platform used to call multiple LLMs through a common interface. Here it is used to run four models in parallel for comparison and generation tasks.
A Gemini model variant that was noted as moving out of preview status.
A Claude model used in the newsletter's example to run Python code and analyze a floor plan. It is discussed as part of an agentic workflow inside Claude Cowork.
Agents that perform coding tasks and can increasingly orchestrate adjacent workflows like design. The newsletter uses them as the execution layer for Design.md scripts.
The class of models discussed as having a blind spot with continuous, high-dimensional, noisy data. This concept is used to frame a limitation in current AI capabilities.
A Qwen model release referenced alongside Qwen3.6-Plus and integrated with opencode. It is one of the named models in the announcement.
A W3C-backed browser extension that exposes website functionality to MCP-capable agents. It lets developers register site functions as structured tools in the browser.
A browser automation protocol used here to let a Claude Code agent control Chrome programmatically.
A Qwen model release with day-0 support for multimodal integration. The newsletter highlights its immediate compatibility with MLX-VLM for visual-language workflows.
Simon Willison’s command-line LLM tool for interacting with models and APIs. This release adds support for OpenAI’s Responses endpoint and better reasoning-token handling.
A workflow for using AI agents to plan, build, test, and update software with minimal manual intervention. The newsletter treats it as a practical product-development paradigm.
A generative media company referenced as an example of a public Discord-based workflow. It is used here to support the idea that visible communities can accelerate learning and product adoption.
Google AI Edge Gallery is a Google tool for showcasing and running on-device AI experiences at the edge, including offline use cases.
A collection of techniques and patterns for building agentic systems. The newsletter frames it as a guide page for AI builders.
A security risk pattern where AI agents have private data access, ingest untrusted content, and can exfiltrate data. For AI PMs, it is a key framework for designing safe agent features.
A Chinese AI lab referenced as releasing GLM-5.2 and publishing open weights. The newsletter cites it as a major open-weights model developer.
Google’s Gemma model family, referenced here as one of the local models run on a Mac. It is part of a broader local-model setup.
A developer and author discussing model behavior and tool-calling reliability. In this newsletter he is cited for analyzing why newer Claude models can produce malformed tool calls.
A test-driven development pattern adapted for coding agents. It emphasizes an iterative failure/success loop that can make agentic coding more reliable.
A Google AI text-to-speech model with native multi-speaker dialogue support across many languages. It is positioned as part of the Gemini product family.
An ecommerce company referenced for its public, Slack-based coding agent River. The example is used to discuss how visible workflows can accelerate learning and adoption.
Technology company that offers the Granite family of models. In this newsletter it appears in relation to Simon Willison's prompting experiments with Granite 4.1 3B.
A quoted individual in a commentary about code quality incentives in AI systems. The newsletter uses him as the source of a viewpoint on maintainable code.
A repository for researching LLM providers' HTTP APIs. It supports abstraction-layer decisions for developers building against multiple model providers.
A product and engineering concept describing the hidden cost of AI-accelerated development when teams lose shared understanding of the system. It reframes debt from code maintenance to team cognition and system comprehension.
OpenAI leader and product/engineering voice associated here with confirming Codex’s unification with the main model. The newsletter cites him via Simon Willison’s note.
A Codex-powered model release from OpenAI aimed at developers and product teams. The newsletter emphasizes its availability as a research preview and its high token throughput.
An open-weight multimodal model in Alibaba's Qwen3.5 series, aimed at agentic and vision-capable use cases. It is relevant to PMs evaluating model capabilities, openness, and deployment options.
A Python library for working with LLM providers through an abstraction layer. The newsletter notes that API research is informing a major change to its provider abstraction.
Stay updated on Simon Willison
Get curated AI PM insights delivered daily — covering this and 1,000+ other sources.
Subscribe Free