↳ Project Finance & Commercial Lenders

Your solar exposure
is growing. Your
visibility into it isn't.

Solar is the #1 driver of criticized loan growth at major project finance banks. The problem isn't defaults — it's that deterioration is invisible until a borrower-prepared report lands four to six months late. LCOE.ai closes that gap, deployed against $250B in active US solar loans.

Market Context · Project Finance Solar
The Scale of the Problem
YoY growth in criticized solar loans+40% (2025)
US solar debt financing, Q1 2026$8.9B — decade high
Avg PV underperformance vs. forecast8.6% (kWh Analytics)
Annual economic leakage — US solar$2.5–5.5B
Typical detection lag4–6 months
LCOE.ai detectionDay of occurrence
Solar loans rarely go to zero. They bleed value through stale data, missed covenants, and late detection — long before default.

The default-loss pool is millions.
The performance-leakage pool is billions.

Lender credit losses after recoveries are bounded — $100–300M annually. The real exposure is the $2.5–5.5B in annual economic leakage across underperformance, debt stress, waivers, and late detection. By the time a borrower's quarterly report arrives, the damage is already months old.

"The borrower's Q2 report looked fine. LCOE.ai showed us actual production was 14% below model — for six months."

Annual Economic Leakage — US Solar
Lender credit losses (net of recoveries)$100–300M
Revenue lost to underperformance & equipment$1.0–3.0B
Debt stress: waivers, sweeps, restructuring lag$250–750M
Tax equity & sponsor production shortfalls$500M–1.5B
Total annual economic leakage$2.5–5.5B
🗓️

4–6 month detection lag

Borrower-prepared quarterly reports arrive late. Covenant drift, DSCR erosion, and production shortfalls compound invisibly in the gap.

📂

No independent verification of borrower data

Each borrower submits in their own format. Manual ingestion is slow, error-prone, and not tied to actual performance data.

🔎

Covenants exist in documents, not systems

DSCR thresholds and reserve requirements live in the loan agreement — but no live system watches them against actual project performance.

Continuous, independent, covenant-aware intelligence on every loan.

Direct integration with 14+ monitoring platforms. The same data the operator sees — at the same time, not at the next quarterly report.

i

Every covenant parsed and monitored.

LCOE.ai reads the loan agreement, LLCA, PPA, and pro forma. Every covenant is mapped to the live data stream that governs it. Breaches surface the day they occur.

  • DSCR, uptime, reserve covenants mapped to live data
  • Notification events triggered automatically
  • Cure period tracking from day one
  • Full loan document ingestion
ii

Live data — independent of the borrower.

Direct API integration with 14+ monitoring platforms. You see the same data the operator sees — independently, in real time.

  • 14+ monitoring platform integrations
  • Production vs. P50/P90 model, updated daily
  • Borrower-reported vs. actual — side by side
  • Format-normalized regardless of submission type
iii

A watchlist that writes itself.

Every loan AI-ranked by emerging risk. Performance drift, DSCR trending, and covenant proximity surfaced continuously.

  • Portfolio-wide risk ranking, updated live
  • DSCR trending and covenant proximity alerts
  • Performance drift flagged before reports arrive
  • Early-action window — weeks before breach
iv

Examiner-ready by default.

A single, time-stamped source of record for every loan. Variance, covenant status, and risk classification continuously documented and exportable.

  • Timestamped covenant monitoring history
  • Borrower report vs. independent data record
  • Risk classification documentation
  • Export-ready for regulatory examination
Illustrative ROI

A regional bank with $2B in solar exposure.

At 8.6% average production underperformance — the industry mean — the leakage exposure on the revenue base alone is material. The cost to monitor it is less than 1 basis point of exposure protected.

$25–50M
Annual leakage
At 8.6% miss on a $2B book's revenue base — the exposure LCOE.ai identifies and helps arrest.
4–6 mo
Detection lag eliminated
Issues surface the day they occur — not the quarter after a borrower-prepared report.
< 1 bps
Cost of monitoring
Less than a single basis point of the exposure protected. ROI compounds across every loan in the book.
📋
Examiner-ready
Time-stamped source of record per loan. Exam prep becomes a download.

Your entire solar book. One view.

Continuous performance monitoring, independent borrower report verification, and covenant intelligence — all in one place.

LCOE.ai · Portfolio Monitor · Lender View
Lending Book
Alerts & Actions
Audit Trail
click tabs · click rows to drill in
12
Projects Monitored
Real-time
2
Report Variances
vs. actual data
3
Covenant Watch
DSCR approaching
0
Missed Events
YTD
↓ Click any project to open full detail view
ProjectPerformance vs. ModelBorrower ReportVarianceDSCR (Live)CovenantStatus
Pecos Solar I — TX−14.2% vs P50−2.1% reported12.1% gap 🔴1.21×0.01× bufferREVIEW
Sunbelt Portfolio — AZ−6.8% vs P50+0.4% reported7.2% gap 🟡1.27×0.07× bufferWATCH
Clearfield Solar — NV+1.4% vs P50+1.2% reported0.2% gap ✓1.44×0.19× bufferOK
Blue Ridge PV — GA−0.8% vs P50−0.9% reported0.1% gap ✓1.38×0.18× bufferOK
Ask AI about your loan book
AI
LCOE.ai Assistant
Connected to your loan book
Hi — I'm your LCOE.ai assistant. Ask me about covenant proximity, borrower reporting variance, DSCR trends, and risk classification across your loan book.
LCOE.ai · grounded in independent monitoring data

30-day pilot. 10 loans. No commitment.

We'll connect to your existing monitoring data and map your covenant package. Most books are live within 30 days.