Western #AI companies: “Protect us from Chinese AI companies!” Also Western AI companies: “NO you can’t use our models for cybersecurity” Hugging Face ended up using Z.ai GLM-5.2 to defend itself from OpenAI rogue models: https://huggingface.co/blog/security-incident-july-2026 Whenever you hear a Western politician talking about banning/regulating open weights models, it’s because a broligarch has his hands…
Tell AI what wins. One of the things that will degrade AI performance the most – and silently – is conflicting instructions where the AI has to determine what you mean. If you look at the reasoning traces (internal chat) when you give conflicting instructions, you’ll see AI models literally chasing their tails. It looks…
_Almost Timely News: 🗞️ What Would A Purely AI-Generated Newsletter Look Like? (2026-08-30) :: View in Browser_ The Big Plug ✍️ Enroll in my new course, AI for Writers and learn how to make AI write better. Content Authenticity Statement 100% of this week’s newsletter was made by me, the human, BUT you will see…
Disclosure: 100% of this document is AI-generated, made by GLM-5.3-Flash on DeepInfra using the OpenCode harness. The silent stretch where your customers go dark Somewhere in your CRM right now sits a customer who bought from you almost a year ago. They opened everything you sent them for the first few weeks after the purchase.…
“Without pulling up the document, tell me what’s in it.” That one sentence tells you in seconds whether or not something is AI slop. You don’t need an AI detector. You don’t need fancy software. You don’t need to burn more tokens or waste more resources. Whether you’re a supervisor, a parent, a teacher, or…
“when everyone’s super… no one will be!” – Syndrome, the Incredibles When everyone has access to state of the art models, your AI is no longer any kind of advantage, and depending on you or your team’s level of skill, might even be a disadvantage. So, if AI isn’t an advantage any more, then what…
This one hastily made slide explains pre-IPO jitters better than anything else when it comes to AI. DeepSeek released its v4 Flash model earlier this year, on April 24. At its release, it was widely regarded as pretty smart and stupid cheap for its size. For a lot of folks using agentic AI, its pay…
Let’s examine this pile of AI slop, generated by ChatGPT: If you’re a writer, you’ve probably noticed something changing fast: AI isn’t replacing the need for great writing. It’s changing how great writers work. The opportunity isn’t to hand your writing over to AI and hope for the best. It’s to learn how to use…
Fifth generation AI models understand time. I discovered this morning, working in Claude Code with Sonnet 5, that it can have an awareness of time and work to a time budget. Previous generations of models lacked this awareness; it seems like models that are aligned strongly for agentic AI can do this. Here’s how it…
