Upstart, a peer-to-peer lending platform which uses AI to make credit decisions and to eliminate fraud, raises $50M Series D, bringing the total raised to $160M
Upstart, a startup founded by ex-Googlers that uses AI to identify who should get a loan and of what size, said today it has raised $50 million.
Context & Ripple Effects
This $50M round extends an arc that began with Upstart's $32.5M Series D in 2017, when the ex-Googler founding team also hired former Google VP Sanjay Datta as CFO to professionalize the finance side. Two years on, total raised reaches $160M — a signal that investors were still funding AI-underwritten consumer lending ahead of any public-market test.
The corpus shows where that bet landed: Upstart later went public with a 47% first-day pop, then saw its stock fall more than half when it cut its 2022 revenue forecast citing rising interest rates. This 2019 raise sits at the private-capital peak of that cycle.
First-order effects
- Upstart gains $50M to scale its AI credit-decisioning and fraud-elimination engine, with $160M total raised giving it runway to grow loan volume without near-term revenue pressure.
Second-order effects
- The round validated AI underwriting as a fundable category beyond Upstart itself — months later, auto-loan underwriter Lendbuzz raised $130M in debt plus $20M in equity led by 83North (Lendbuzz's debt-and-equity raise), showing the model spreading to adjacent verticals.
Third-order effects
- Upstart's own later trajectory — the 56% single-day drop when rising rates forced a revenue-guidance cut — reveals the structural exposure of algorithmic lenders: models trained on cheap-credit-era data reprice badly when the rate cycle turns, making AI credit platforms more macro-sensitive than their fraud-and-default metrics suggest.
The trend: Venture capital is underwriting AI credit-decisioning startups across consumer-lending verticals, but the category's economics remain hostage to interest-rate cycles rather than model quality alone.