Cost a scripted run before you build it
Lesson 3 of 5 · The teacher's week
1 · Learn the move · Comparison matrix
The moment a teacher thinks 'I could script this' is the moment to stop and cost it, because the script that runs on the whole grade's data is the script that runs on the whole grade's names. Build the matrix before the script: what goes in (and whether a student is in it by name or number), what the tool does with it, who has to approve the tool for that input, how many runs a year, what a wrong run costs (a wrong worksheet is a Tuesday; a wrong letter home is a meeting; a student record in an unapproved tool is a report to the district), and what the by-hand cost is for comparison. Most scripted runs on this door never clear the first column: they need the roster, and the roster does not go in. The ones that do clear it (a worksheet set from a standard, a weekly reading list, a sub-plan folder) are worth scripting. The decision to build stays yours, and it is made with the matrix, not the enthusiasm.
I am thinking of scripting this task: [describe it]. Build a cost matrix before I build anything: WHAT GOES IN (and whether a student is in it by name, by number, or not at all); WHAT THE TOOL DOES WITH IT; WHO APPROVES THE TOOL FOR THAT INPUT at my district; RUNS PER YEAR; COST OF A WRONG RUN, in the words of what actually happens; BY-HAND COST. Fill each cell from what I told you or write [YOUR CALL: the fact]. End with the one column that decides it. Do not recommend building; show the matrix.
Vendor notes, dated
Prices and plan names move. Each note says when it was looked up, whether a person has confirmed it, and the day it comes due. Past that day, treat it as a question.
- Claude. Claude Pro is $20 a month billed monthly, or $17 a month billed annually. Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- ChatGPT. ChatGPT Plus is $20 a month. Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- Google Gemini. Google Gemini's paid consumer plan is Google AI Pro at $19.99 a month; Google AI Ultra starts at $99.99 a month. Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- Anthropic. Anthropic's consumer plans (Free, Pro, Max) may use chats to train models under a setting the user controls, with five-year retention when training is on and 30 days when it is off; Claude for Work, Education, Government and the API are excluded. [P] Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- Anthropic. Anthropic's usage policy, effective 2025-09-15, requires a qualified human reviewer and disclosure of AI use for housing, legal and financial decisions. [P] Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- OpenAI. OpenAI's consumer plans train on chats unless the Improve the model setting is turned off, and a Temporary Chat is not used for training; Enterprise, Business, Edu and API data are not used by default. [S] Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- Google Gemini. Google Gemini Apps keeps activity 18 months by default, keeps human-reviewed chats up to three years, and tells users not to enter confidential information. [P] Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- Microsoft Copilot. Microsoft Copilot on a personal account may use chats to train models unless the user opts out; Entra and Microsoft 365 work accounts are excluded. [S] Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- Claude. Claude Team, Enterprise and API inputs and outputs are not used for training by default; a BAA is available only on Enterprise and the first-party API after admin action. Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- Claude. Claude Free, Pro and Max inputs can be used for training if the user allows model improvement, with retention up to five years if allowed and 30-day deletion otherwise (pages dated 2026-03-16 and 2026-07-01). Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- OpenAI. OpenAI API inputs are not used for training by default; abuse-monitoring logs are kept up to 30 days; zero data retention is available by approval. Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- OpenAI. OpenAI Business, Enterprise and Edu no-training defaults are reported by secondary sources only; the vendor trust pages returned 403 on the check date. Read the vendor page yourself. Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- Microsoft. Microsoft Copilot and Copilot Chat with Enterprise Data Protection: prompts and responses are not used to train foundation models, the DPA applies, and Microsoft acts as processor (Learn page updated 2026-08-18). Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- Google. Gemini in Google Workspace content is not human-reviewed or used for training outside the customer's domain; default retention is 18 months. Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
- Every vendor. Every vendor's no-training or no-retention term is a contract term, not the determination: it does not by itself make the vendor a school official with a legitimate educational interest under 34 C.F.R. 99.31(a)(1), which the district makes in its own agreement, and it says nothing about whether the plan a teacher actually uses is the plan the term covers. Researched Oct 6, 2026, not yet confirmed by a person, recheck by Dec 15, 2026.
2 · Your turn. You write the prompt
A Ridgeway High teacher wants to script the weekly progress emails to the families of students who are missing work. The matrix stops it in the first cell: the input is the gradebook export, which carries every student by name and number, into a tool the district has not approved for records. A wrong run is a letter to the wrong family. The by-hand cost is forty minutes a week. She scripts the worksheet set from the week's standard instead, which needs no student at all, runs weekly, and costs a Tuesday when it is wrong.
Write the prompt for this scenario as given, or swap in a real one from your own work. The score works the same either way.
Remember: the AI sees only your prompt, not this page. If the situation isn't in your prompt, it doesn't exist.
Optional. These shape the output when you run your prompt below, not your score.