HeyJaredPurposeful AI
Partner for Higher Education and Non-Profit organizations navigating the dynamic opportunities of Artificial Intelligence.
- Indexed issues, last 90 days
- 13
- Latest publication
- Sep 29, 2026
- Audience
- Checking…
- Earliest in this view
- Jul 5, 2026
Latest issues
The Question That Comes First (opens the original)
Read excerpt
I’ve been asked, “What AI system should I use?” several hundred times now, and I’ve never once given an answer that helped.I do have preferences. The trouble is the shape of the question, which makes every available answer useless. Claude or Gemini or the other one. Which is best. The honest response is either “all of them” or “none of them,” depending on the week. That sounds evasive until you look at what the question takes for granted: that the skill lives inside the product, that there’s a r
The Middle Is Emptying Out (opens the original)
Read excerpt
The most interesting people in any room about AI are the moderate skeptics.The enthusiasts show up already convinced and mostly want to tell you about their workflow. The hardliners have decided the whole thing is a plagiarism machine wearing a lab coat, and they didn’t come to be moved. The moderate skeptics have real reservations, they argue them in good faith, and they came anyway. They’re the ones worth watching, because something happens to them that doesn’t happen to anyone else in the roo
Make It Refuse You (opens the original)
Read excerpt
I want to start with something that sounds like a complaint about consulting and is not one.Every client request pushes my thinking at an angle I would not have chosen. Not a better angle, per se, just a foreign one. Whenever I am working with a client, the problem arrives already shaped by constraints belonging to someone else: an institution’s calendar, an accreditor’s expectations, a political situation inside a department that I only half understand. I could not have generated that shape sit
The Workflow Design Bot (opens the original)
Read excerpt
This morning I was working in Claude Code and something felt off.Not wrong. The code was getting written. Files were changing. Tests were passing. But I kept losing the thread of which files had been touched and whether the thing taking shape on screen matched the thing I set out to build. At one point I read a summary of what had just happened and realized I was reconstructing my own project from a report.There is a name for that sensation in the human-factors literature. Work on shared mental
Obstruct the Machine (opens the original)
Read excerpt
There’s a version of AI literacy going around that sounds responsible and is actually hollow. It goes like this: you need a better prompt, and you need a better tool. Learn the prompt patterns. Upgrade to the model with the bigger number. Add the integration. Then you’ll be using AI well.Take that advice apart and it says two different things. To someone actually tuning a system, a prompt is an input you revise and a tool is a model you swap. Both are settings, and the claim is that better setti
Publishing over time
Last 90 days. Choose a month to open its work.
Recurring subjects
Named in the text we hold. One piece can cover several.
Audience
No verified audience measurement yet.
About this data
Counts cover the work we have indexed. Tone needs enough text and a confident classification. Excerpts and episode notes are not full articles or transcripts.
Identity or attribution wrong? Suggest a correction.