Sharing AI-Usable Public Comments for NegReg
If you want to leverage 1,100+ comments before the Pell Grant discussions next week

Was this forwarded to you by a friend? Sign up, and get your own copy of the news that matters sent to your inbox every week. Sign up for the On EdTech newsletter. Interested in additional analysis? Upgrade to the On EdTech+ newsletter.
Next week the US Department of Education (ED) begins the first of two weeks for the Accountability in Higher Education and Access through Demand-driven Workforce Pell (AHEAD) negotiated rulemaking (NegReg) committee that is tasked with implementing many of the most important provisions in the OBBBA bill signed this summer. The December 8-12 meetings are focused on changes to the changes to Pell Grants, including the new Workforce Pell creation.
These are big changes, and the details of the rules will have a significant impact on US higher education. Accordingly, there have been quite a few recent posts with groups advocating their positions and explaining what is at stake.
In the spirit of improving transparency, I would like to extend my efforts at analyzing the public comments submitted this summer and share with the community.
The Department of Education (ED) is now preparing to define the regulations and guidance required to implement the OBBBA provisions, and the first step was gathering public comments. Thanks to my research assistants ChatGPT and NotebookLM, I have pulled together all 1,124 public comments (including the text from PDF attachments) into this summary (you can find my initial analysis in this post). You can think of this as AI-enabled crowdsourcing of what ED should consider in its rulemaking, and this should provide a guide on the topics to be covered in negotiated rulemaking (aka negreg).
The Files
I am sharing below are two files:
A csv file including all of the comments, which can be directly used in ChatGPT.
A txt version of that csv file, which can be directly used in NotebookLM.
The format for each is a spreadsheet with the Document ID, Posted Date, Name of submitter, Organization of submitter, the text Comment of the submission, the Category of the Organization, and the pdf_text where I extracted the comments from attached PDFs and embedded directly in the spreadsheet (in just over a third of cases, the primary comments were in the attachment, not in the direct submission).

ChatGPT Usage
Anyone can attach the csv file to ChatGPT and then have a go at it. For example, here I prompt “What are the top 10 asks regarding Workforce Pell? Focus on safeguards needed and implementation issues that would help the new rules have the intended impact with minimal unintended consequences. Do this in the style of a Dave Letter Top 10 list.”
( Cue the drum roll, Paul… )
💥 The Top 10 Asks for Workforce Pell Implementation (Letterman-Style) 💥
10. “Require programs to prove they actually lead to jobs… not just lead to another press release.”
Make job placement data real, verifiable, and transparent. No more “98% employed!” claims where half the graduates work at the founder’s brother-in-law’s vape shop.
9. “Make sure short-term programs don’t become the Wild West of federal aid.”
Clock-hour chaos, anyone? Workforce Pell should not be the new frontier where shady bootcamps plant their flag and yell, “It’s free money, baby!”
8. “Set a minimum bar for program quality that’s higher than ‘we have a pulse.’”
Minimum instructional hours, qualified instructors, and real curriculum oversight. Not the “one-person with a PowerPoint deck” model.
7. “Ensure institutions can’t magically re-brand existing low-performing programs as ‘workforce.’”
Slapping the word tech or certificate on a failing program doesn’t suddenly make it worthy of federal dollars. (Nice try, though.)
6. “Protect students from programs whose tuition-to-wage ratio screams ‘bad life choices.’”
Require transparent cost, earnings, and repayment expectations—before enrollment. Because nobody wants a bill bigger than their paycheck.
5. “Tie eligibility to real employer validation—not just a letter signed by the HR intern.”
Employer partnerships should show actual hiring demand, regional relevance, and a track record of placements. No photocopied ‘support’ letters allowed.
4. “Implement a strong, simple ‘Do No Harm’ test for short-term programs.”
If programs leave graduates financially worse off, they’re out. Call it the “Don’t-ruin-lives Pell safeguard,” trademark pending.
3. “Guarantee strong state and accreditor accountability so no one plays the ‘not our job!’ game.”
If something goes wrong, we shouldn’t need a whiteboard to figure out who's responsible. States, accreditors, and ED: pick up the phone.
2. “Require clear reporting so policymakers can actually monitor outcomes… not guess them.”
Common metrics. Annual submissions. Public dashboards. No 400-page PDFs updated every five years where page 237 holds the only number that matters.
And the Number One Ask for Workforce Pell Implementation…”
1. “Stop bad actors before they start—don’t wait for a Netflix documentary to clean up the mess.”
ED should use early-warning indicators, financial monitoring, and rapid reviews. If a provider looks like a scandal in progress… it probably is.
NotebookLM Usage
Anyone can attach the txt file to NotebookLM and then have a go at it. For example, here is an infographic of the Top 10 asks with the same description as above, without the David Letter style.

Purpose
These are merely examples, and you can do your own AI-assisted research. If the results from this data share make it into the NegReg discussions next week, speak kindly of me. Better yet, in the old blogging style, share you results in public.
The main On EdTech newsletter is free to share in part or in whole. All we ask is attribution.
Thanks for being a subscriber.
