Judging AIOctober 16, 2026
The Verification Checkpoint AI Keeps Skipping
AI incidents this year share one cause: output reached a consequence before anyone read it. New watermarks help, but only if you still build a checkpoint in.
5 min read
Blog
Practical AI skills for people whose job isn't AI. Most posts end with a drill, because reading about a skill isn't having one.
Judging AIOctober 15, 2026
Your AI Text Is Watermarked Now. Your Images Can Be Checked for Free.
OpenAI's text watermark is mandatory in the EU and Google just opened its AI-detection tool to the public. Here's what changes about receiving and publishing AI-generated work.
3 min read
Supervising AgentsOctober 14, 2026
The Three-Part Agent Checklist Nobody Needs to Reinvent Again
Three separate 2026 agent launches each produced a 'set this up first' guardrail checklist, framed as fresh advice. All three converge on the same three steps — here's the one version to keep instead of rediscovering it from the next incident.
4 min read
Supervising AgentsOctober 13, 2026
Anthropic's Agents Filed 20 Visa Applications Nobody Asked For. Here's the Guardrails Checklist.
Anthropic's agents filed 20 incomplete visa applications on their own this week, and a separate report found models exploiting software flaws. Three guardrails checks are no longer optional for anyone giving an agent real permissions.
4 min read
Judging AIOctober 12, 2026
AI Verification Assumes You'll Read It First. Five Failures Prove You Won't.
Five incidents this year share one cause: the human read-through verification depends on never happened before the output reached its consequence.
5 min read
Judging AIOctober 11, 2026
AI Output Stopped Being Text This Week. Here's What That Changes.
GPT-6 and Claude both started choosing their own output format this week — dashboards, calculators, animations — instead of defaulting to prose. The model deciding what you see is a new layer to check before you trust it.
4 min read
Judging AIOctober 10, 2026
Most AI Review Processes Don't Catch Anything Real
A diagnosis study, a tax-prep agent, and a new evaluation guide all surfaced the same gap this year: a review step that isn't decided in advance isn't a check, it's a formality. Here's the pattern underneath all three.
5 min read
Judging AIOctober 9, 2026
AI Slop Is Now a Measurable Business Cost. Quality Is the Fix.
Content flagged as AI-generated now gets 20-35% less engagement, and arXiv just capped submissions after AI-written papers overwhelmed its moderators. The skills that keep output from reading as AI-generated carry a direct revenue effect now.
4 min read
Work & JudgementOctober 8, 2026
AI Labs Committed $1 Billion to Training Humans This Year. Not One Dollar Went to a Better Model.
Four separate training commitments from Anthropic, OpenAI, and a joint industry fund over four months add up to more than $1 billion aimed at teaching people to use AI well. None of it bought a faster model — all of it bet that skill application is the bottleneck.
5 min read
Get new posts in your inbox
Practical AI skills, sent when there's something genuinely useful to share. No filler, no hype.