The Gate That Approved the Broken Draft
A new meaning-check flagged two clean AI drafts and missed two broken ones. Here’s what that reveals about trusting an automated writing gate.
The Script That Couldn’t See Free Models
My own tool hid every free model on OpenRouter, including one that turned out to be GLM-5.3-Flash. Free changed nothing about how I judge a draft.
The Second Model Marked the Fiction as Sourced
I tested whether a second AI model can fact-check the first one’s draft against the source. It caught 4 of 10 fabrications, and once it noted the mismatch and still marked the sentence sourced.
The Gate Nobody Walked Through
I built a safety check for AI delegation, rolled it out to a second site, and gave it two real jobs. Neither one used it. Here is what that tells me.
I Was Scoring the Wrong Thing
I gave four cheap models a smaller job: reorganize my research notes. Invented figures went to zero. Two of the four deleted the caveats instead.
Cheap Models Are Cheap. Reviewing Their Output Is Not.
I scored eleven AI-written drafts of the same article, from $0.003 to $0.10 each. All eleven failed the same item, and paying ten times more did not raise the score.
I Stopped Writing Rules for My AI Agents. I Wrote Gates Instead.
After four incidents, I stopped trusting written rules alone and started wiring Claude Code hooks into scripts that block the risky action outright.
How I Spent 3 Days of Frontier-Model Access Auditing My AI Stack
I had 3 days of full access to a new frontier model before my subscription capacity reset. Here’s why I used it to audit my own AI automation system instead of building something new, and what it found.
How I Fact-Check AI-Written Drafts Before Publishing
The biggest upgrade to my AI fact-checking wasn’t a smarter model or a longer prompt — it was switching the context. Here’s the simple move that catches what an AI’s self-check misses.
My WordPress Publishing Flow: Draft → Scheduled → Synced
Scheduling a WordPress post isn’t the finish line. Here’s my Draft, Scheduled, Synced flow, the post-publish check most people skip, and the two silent failures that taught me to close the loop.