Aaron Chartier — a technologist's workshop: tools, agents, and playful machines.https://aaronchartier.com/en-ushttps://aaronchartier.com/blog/perplexity-won-the-wrong-argument/https://aaronchartier.com/blog/perplexity-won-the-wrong-argument/The Ninth Circuit vacated Amazon's injunction against Comet: a shopping agent is a tool, not a person under the CFAA. The strategy that won this round is the one that loses everything that comes after it.Wed, 05 Aug 2026 00:00:00 GMTai-agentslawcfaaweb-infrastructurehttps://aaronchartier.com/blog/provenance-proves-the-build-not-the-builder/https://aaronchartier.com/blog/provenance-proves-the-build-not-the-builder/npm's own trusted-publishing pipeline cryptographically attested a worm on August 4 because the compromise happened one layer above the token it was designed to protect. The same gap just showed up in agent skill marketplaces and MCP registries too.Tue, 04 Aug 2026 00:00:00 GMTsupply-chainnpmagentssecurityhttps://aaronchartier.com/blog/cheaper-tokens-richer-labs/https://aaronchartier.com/blog/cheaper-tokens-richer-labs/OpenAI just cut its cheapest model's price 80%. Anthropic's inference margins climbed from 38% to over 70% in the same year. Both are true at once, and the reason says more about who holds pricing power in AI than any per-token chart does.Sat, 01 Aug 2026 00:00:00 GMTaieconomicsinferencehttps://aaronchartier.com/blog/multi-agent-systems-spend-more-and-call-it-architecture/https://aaronchartier.com/blog/multi-agent-systems-spend-more-and-call-it-architecture/Two 2026 papers show automated multi-agent LLM systems losing to a single agent at matched compute — and the exception, Anthropic's own research system, turns out to prove the same rule from the other side.Fri, 31 Jul 2026 00:00:00 GMTagentsllm-systemsdistributed-systemshttps://aaronchartier.com/blog/the-weights-didnt-change-the-score-did/https://aaronchartier.com/blog/the-weights-didnt-change-the-score-did/Anthropic revised a prior model's benchmark score in May 2026; OpenAI nearly tripled another's two months later — both without touching the weights, only the harness. What that says about self-run evals.Thu, 30 Jul 2026 00:00:00 GMTbenchmarksevalsaihttps://aaronchartier.com/blog/agent-memory-solves-storage-not-recall/https://aaronchartier.com/blog/agent-memory-solves-storage-not-recall/Cloudflare and MinIO both shipped durable 'agent memory' products this year, framed as the fix for context rot. The research the fix is named after says the hard part was never storage.Wed, 29 Jul 2026 00:00:00 GMTllmsmemorycontext-engineeringhttps://aaronchartier.com/blog/mcp-remote-code-execution-design-decision/https://aaronchartier.com/blog/mcp-remote-code-execution-design-decision/OX Security's April disclosure showed every official MCP SDK will run an unsanitized command string as a subprocess. The answer was a documentation update, because on the transport's own terms this is intended behaviour — and neutral governance has no way to overrule an argument of that shape.Tue, 28 Jul 2026 00:00:00 GMTmcpsecurityagentshttps://aaronchartier.com/blog/kimi-k3-license-open-weight-went-political/https://aaronchartier.com/blog/kimi-k3-license-open-weight-went-political/The letter 25 companies signed in July 2026 argued open and closed models have converged enough that regulators should back off. Days later, the open model closest to proving that point shipped a tighter license than its predecessor.Mon, 27 Jul 2026 00:00:00 GMTailicensingopen-sourcehttps://aaronchartier.com/blog/the-crawler-rulebook-finally-has-an-enforcer/https://aaronchartier.com/blog/the-crawler-rulebook-finally-has-an-enforcer/Cloudflare will start blocking mixed-use AI crawlers by default this September. The mechanism that makes it possible is cryptographic, but the power it hands out isn't going to a standards body.Fri, 24 Jul 2026 00:00:00 GMTaicrawlersinfrastructureeconomicshttps://aaronchartier.com/blog/the-queue-got-longer-the-shelf-didnt/https://aaronchartier.com/blog/the-queue-got-longer-the-shelf-didnt/App Store submissions are up 80% year over year on iOS alone, and Apple just spent a guideline update policing apps that make other apps. The bottleneck was never the building.Wed, 22 Jul 2026 00:00:00 GMTapp-storedistributionsoftware-economicshttps://aaronchartier.com/blog/valid-json-was-never-the-hard-part/https://aaronchartier.com/blog/valid-json-was-never-the-hard-part/Grammar-constrained decoding guarantees a language model's output will parse. Benchmarks from 2024 through 2026 keep finding that guarantee stops well short of the output being right.Mon, 20 Jul 2026 00:00:00 GMTllmstructured-outputconstrained-decodinghttps://aaronchartier.com/blog/the-bug-reports-got-better-and-thats-the-problem/https://aaronchartier.com/blog/the-bug-reports-got-better-and-thats-the-problem/Curl killed its bug bounty in January over AI slop. By April the slop was gone and maintainers had more work than ever — because the thing they were filtering for was never correctness.Sat, 18 Jul 2026 00:00:00 GMTopen-sourcecode-reviewaisecurityhttps://aaronchartier.com/blog/the-browser-finally-speaks-gguf/https://aaronchartier.com/blog/the-browser-finally-speaks-gguf/llama.cpp shipped a real WebGPU backend, benchmarked this May against WebLLM and Transformers.js. The in-browser model catalog just got 400x bigger — and the numbers show exactly what that gain cost.Fri, 17 Jul 2026 00:00:00 GMTwebgpuon-device-aillama.cpphttps://aaronchartier.com/blog/the-api-key-was-always-the-weird-part/https://aaronchartier.com/blog/the-api-key-was-always-the-weird-part/Consumer AI subscriptions are starting to authenticate the apps you use — no developer key, no token bill. It's a real shift in who holds the meter, and it lasts exactly as long as the labs allow it.Thu, 16 Jul 2026 00:00:00 GMTagentsaieconomicshttps://aaronchartier.com/blog/you-cant-scale-what-you-have-to-remember/https://aaronchartier.com/blog/you-cant-scale-what-you-have-to-remember/MCP is dropping the session. That reads as a loss until you remember why every protocol that mattered ended up stateless — and what it actually costs to get there.Wed, 15 Jul 2026 00:00:00 GMTmcpagentsweb-architecturehttps://aaronchartier.com/blog/the-control-question-already-has-an-answer/https://aaronchartier.com/blog/the-control-question-already-has-an-answer/The argument over whether capable AI systems could be kept in check is effectively over. It wasn't won — the careful people just stopped holding the line, and an unattended ransomware agent showed what that costs.Wed, 08 Jul 2026 00:00:00 GMTagentssecuritytrusthttps://aaronchartier.com/blog/nobody-is-reading-your-llms-txt/https://aaronchartier.com/blog/nobody-is-reading-your-llms-txt/97% of llms.txt files never get read, and we've run this experiment before. What separates the sidecar files that shaped the web from the ones that became decoration.Fri, 03 Jul 2026 00:00:00 GMTagentsllms-txtweb-history