miles dinsmore

First real post is up on milesdinsmore.com I wrote about Docker Sandboxes, safer AI coding workflows, and why I think agents need guardrails before touching real systems. Still learning, still breaking things—just preferably in a sandbox. milesdinsmore.com

@KylerJohnsonDev @itsjustcornbro @ThePrimeagen @pmarca LLMs were trained on high quality writing, which should surprise no one. Don’t let them stop you, AI is imitating you because you’re the blueprint.

I had a second major in college in English Literature. I wrote so many papers thinking about it makes my head spin. But I liked to use the em dash a lot and that habit stuck. Now everyone thinks my responses are just AI generated. It’s pretty lame. So I guess I’m going to have to stop using the em dash…

🚨 NEW RESEARCH: “Lingua Ex Machina: A Procedural Xenolinguistics Engine Reveals Zero-Shot Language Acquisition, Human-Unreadable Coding Systems, and Exploitable Covert Channels in Frontier AI” Some of you may remember the name of this lil engine: GLOSSOPETRAE 👅🪨 Well, we've got upgrades 😎 It started as a procedural xenolinguistics engine: one seed in, an entire alien language out. Phonology, morphology, syntax, writing systems, lexicons, grammar docs, all generated from scratch and internally consistent. Every seed produces a unique language. Every language is deterministic. Then we used it to ask a weirder question: Can frontier AI models use languages that never existed before for practical applications? As it turns out: yes!! They can read them, write them, translate them, code in them, and even use the weird blind spots between tokenizers as covert channels. So this paper explores three ideas at once: ▶️ zero-shot language acquisition ▶️ human-unreadable code that models can still execute ▶️ exploitable covert channels in frontier AI systems GLOSSOPETRAE is no longer just a language generator... 🧵

There's a strange inversion happening with trust. We used to trust institutions and distrust individuals. Now we trust individuals and distrust institutions. This is going to get more extreme as AI floods the internet with corporate sounding content and brands that all look the same because they're all trained on the same 200 references. The more polished and institutional something looks, the less people trust it because it pattern matches to "generated". The more raw, personal, and imperfect something feels, the more people trust it because it pattern matches to "real". I'm noticing this. Maybe you are too.

@xJacksonBarnes You build your own solution or did you find a good CRM with agentic support for your size?

Etsy customer service runs on a bot now. Shipping questions. Refund requests. Order updates. I used to check messages 50 times a day. Now I check once. The win wasn't speed. It was getting my brain back.