What changed
Two now-final Education Department rules (published 2026-05-19 and 2026-07-01 in the Federal Register) create (a) a live Workforce Pell funding stream for short-term programs and (b) the STATS transparency system with an earnings-accountability framework that limits Direct Loan eligibility to programs whose graduates meet earnings benchmarks. FACT per the two cited rule titles/URLs.
Why now
Both rules are final (not proposed), so the compliance and eligibility regime is real and dated; the Workforce Pell stream is being stood up now, meaning providers are actively deciding which short-term programs to launch under a benchmark that determines whether the program keeps its aid eligibility — a decision window that exists only at program-design time.
Converging signals
regulation (STATS earnings accountability, doc 2026-13286) + money (Workforce Pell funding, doc 2026-10013) + ai (cheaper frontier-model inference lowering per-program modeling cost). The rule + the funded program + the modeling capability meet at one point: a pre-launch go/no-go forecast. The AI-cost signal (GPT-5.6 marketing page) is the weakest leg — it is a capability claim, not a demand signal, and the modeling here is mostly BLS/wage data math, not LLM-dependent.
Customer pain
HYPOTHESIS (not evidenced in input — demand_evidence is empty): a provider that launches a short-term program which later misses the earnings benchmark loses aid/loan eligibility and thus the program's revenue line, after sinking curriculum, marketing, and staffing cost. The pain is plausible and structurally forced by the rule, but there is zero complaint/hiring evidence in this input to confirm providers feel it or will pay for foresight.
Who pays
Workforce/short-term training providers (community colleges, career schools, bootcamps, CTE providers) and new Workforce Pell entrants; secondarily the consultants/associations who advise them. The BENEFICIARY (provider keeping eligibility) is also the BUYER here, which is clean.
Solved today
Providers currently guess, use in-house institutional-research staff, or hire higher-ed compliance consultants; occupation-region wage baselines are assembled manually from BLS/state wage data. INFERENCE — not documented in input.
Why current solutions are bad
Manual, slow, and done program-by-program without a repeatable benchmark-math model; consultants are expensive per engagement. INFERENCE.
Proposed product
An occupation × region earnings-baseline dataset (public BLS/state wage + program-outcome data) wrapped in a scoring layer that maps a provider's proposed program (occupation, region, cohort size, expected completion) to the rule's benchmark methodology and outputs a pass-probability with risk drivers. Delivered as a per-program report plus a portfolio-triage monitoring subscription.
MVP version
Ingest BLS OES + relevant state wage data into an occupation-region earnings baseline; hard-code the final rule's benchmark math (thresholds + measurement window); a thin scoring layer that returns pass-probability + top risk drivers for a submitted program. LLM only for narrative risk explanation, not the core math.
30-day build
Read BOTH final rules end-to-end and extract the exact benchmark methodology (threshold definition, occupation/region granularity, earnings measurement window, cohort rules). Backtest the model against any already-published program outcomes (e.g. prior Gainful Employment / College Scorecard earnings data) to prove it beats naive guessing. If methodology is too vague or outcomes too noisy — KILL.
60-day build
Build the baseline dataset + scoring engine; produce sample reports for 8-10 real published programs; take them to 5 pilot providers/consultants for validation and pricing.
90-day revenue plan
Convert 3-5 pilots to paid per-program reports and/or a monitoring subscription; first revenue realistic in this window given the founder can self-fund the build.
Distribution path
Direct outreach to workforce/CTE provider associations, community-college institutional-research offices, career-school compliance staff, and higher-ed compliance consultants; content on 'will your Workforce Pell program pass the earnings gate' targeted at the decision window.
Pricing hypothesis
HYPOTHESIS: $500-$2,000 per program forecast report; $300-$1,000/mo portfolio monitoring subscription. Undercuts a per-engagement consultant.
Technical difficulty
Moderate. The hard part is faithfully encoding the benchmark methodology and validating the model, not the AI — data assembly + benchmark math dominate.
Legal / regulatory risk
Moderate: this is a predictive/advisory product about a federal eligibility rule. Must not represent forecasts as guarantees or official determinations; disclaimer discipline required. No licensure needed to publish a data/forecast product.
Platform dependency
Low — no app-store or platform owner. Depends on continued availability of BLS/wage/outcome data and stability of the rule methodology.
Founder fit
Strong on shape (regulation + public money + public records + data product), which is the founder's primary thesis and edge. Weaker than his FMCSA pattern in one respect: this is a FORECAST/advisory product, not a per-filing submission into a government portal, so the forced-buyer 'must file by deadline' lever is softer — the provider can decline to buy foresight and just launch anyway.
Breakout potential
Good: 50-state replication of wage baselines, expansion into ongoing benchmark monitoring, and adjacency into the actual Workforce Pell / STATS reporting/filing layer (which IS the founder's per-filing sweet spot) once relationships exist.
Final recommendation
WORTH A SCOPED VALIDATION SPRINT, not an immediate build. The regulatory + public-money convergence is real and dated, and the shape fits the founder — but two gates must pass first: (1) read both final rules and confirm the earnings benchmark is concretely defined; (2) backtest against published outcomes to prove the forecast beats naive guessing; and (3) find ANY real demand signal that providers pay for pre-launch foresight. Proceed only if all three hold.
Next action
Pull and read Federal Register docs 2026-13286 (STATS/earnings accountability) and 2026-10013 (Workforce Pell); extract the exact benchmark methodology and measurement window, then backtest the threshold math against existing published program earnings outcomes to confirm it materially beats naive guessing.