← All investments

Vertical Fintech · June 2026

The AI Spend Layer

AI spend — the dollars and the hours a person or company puts into AI tools — is the most predictive underwriting signal of the next decade. The card is the cheapest way to capture it.

v2 Investment Thesis Chinat Yu

TL;DR

AI spend is observable in real time, hard to fake, correlated with future income and survival, and not yet captured by any incumbent bank or card network. The vertical card playbook the best fintech investors already know (SoFi, Figure, the Bill.com pattern) extends cleanly into this category, with the biggest unclaimed positions sitting in consumer-side AI fintech — student-and-young-builder credit cards underwritten on AI-tool spend and engagement.

The thesis points to two new positions worth opening in 2026: one consumer card targeting the student-and-builder population, and one data-layer bet that aggregates AI-spend signal across cards. The two compound — the card supplies the dataset; the data layer monetizes it across banks, employers, and AI vendors.

The Macro Insight

AI-spend depth — both dollars and hours — is the leading indicator of who will produce economic value in the next decade. The institution that captures it first owns the underwriting layer of the post-AI economy.

Three observations make this insight non-consensus.

First, AI-spend depth predicts speed of building. Developers using Copilot complete tasks 55% faster than non-users. The gap widens at the frontier: the people running multiple LLM sessions in parallel, layering Cursor + Claude Code + Replicate + n8n into a single workflow, ship prototypes in hours that used to take quarters. The depth of someone's AI-tool stack is a real-time proxy for how fast they can build, ship, and learn — which is the real input to future income, more than degree or employer.

Second, the signal is concentrated in transaction data. Time-on-tool is hard to observe from outside. Subscription depth is not. Every AI tool worth using sits behind a paywall: $20/mo ChatGPT Plus, $20/mo Claude Pro, $20/mo Cursor, $200/mo Claude Max, $200/mo ChatGPT Pro, $200/mo Cluely. Spend pattern reveals depth. A 19-year-old paying $400/mo across five AI tools out of summer-job money is a stronger forward signal than a 24-year-old with a $90K starting salary and no AI subscriptions. No card today surfaces that comparison.

Third, nobody is underwriting on it. Brex and Ramp see corporate expense data but do not segment AI. SoFi sees student-loan repayment but not tool usage. Greenlight sees teen spend but not AI-tool category. Cherry sees elective-healthcare repeat-spend but not tech. The category is empty.

The thesis in one sentence: AI spend is the new FICO for the under-30 builder generation, and the cheapest way to capture it is by issuing them a card.

Why Now

Three shifts converged in 2025–2026 and make this fundable for the first time.

1. AI subscriptions hit consumer mass. OpenAI passed 400M weekly active users by February 2025 and roughly 900M by February 2026 — a 2.25× expansion in twelve months. Cluely and similar AI-study-and-work assistants have made the under-25 cohort the heaviest AI-tool stackers per capita. The customer base for a consumer AI fintech now exists at scale and is concentrated in identifiable populations.

2. Real-time alt-data underwriting became standard. Affirm built BNPL on permissioned data via Spark and Databricks. Cherry built medical financing on alternative income-cadence signals, achieving 80–90% approval at a fraction of CareCredit's MDR. Karat Financial proved you can underwrite a previously-uncreditable population (creators) on platform earnings rather than W-2. The infra is mature. Underwriting on AI-spend pattern is technically possible in a way it was not in 2023.

3. Card issuance compressed from years to weeks. Lithic ships a sponsor-bank-backed card in three weeks. Stripe Issuing, Marqeta, Highnote, and Treasury Prime have made the BIN partner layer effectively a commodity. The cost of testing a vertical credit card thesis dropped from $5–10M to $200–500K.

These three shifts mean a category that was structurally impossible to enter in 2023 is now economically attractive in 2026 — and will be obvious to every other fund by 2028. The 12–18 month window to take the first conviction position is open right now.

Three Sub-Theses

The category breaks cleanly into three populations. Each has different unit economics. The existing fintech-VC portfolio landscape covers one well, one partially, and one not at all.

Sub-thesis A · B2B

Corporate AI-spend cards

A vertical corporate card that captures, categorizes, and benchmarks AI spend across SaaS tools, then underwrites credit on top of the resulting dataset.

  • Pattern: Brex × Bill.com × Cherry. Vertical card with a data layer that sells back to the segment it serves.
  • Buyer: AI-native companies, seed → Series C. VC-backed CFOs expect median AI tool budgets to double from $20K to $50K in 2026.
  • Unit economics: interchange ~2.0% blended + SaaS dashboard ($200–500/mo) + benchmark data layer sold to enterprises in year 2+.
  • Status: a small number of early-stage entrants. Otherwise empty.
  • Recommended position: monitor. Most obvious category to enter, therefore most contested by 2027. Wait for traction proof on the wedge before leading.

Sub-thesis B · B2C · Largest Unclaimed Position

Consumer AI-spend cards (student-and-builder cohort)

A consumer credit card or charge card targeted at the heaviest per-capita AI users in the population: students, recent graduates, early-career builders, and self-taught programmers. Underwritten not on FICO but on AI-tool engagement and subscription depth.

  • Pattern: Greenlight × SoFi × Karat. A card for a coherent under-served population with non-traditional income, underwritten on alternative-data signals the incumbents do not see.
  • Buyer: ~19M US college students (Fall 2024 enrollment, per the National Student Clearinghouse Research Center) plus ~8M post-grad early-career builders aged 22–28. Penetration of $20+/mo paid AI tools in this cohort is rising fast; median spend across the heaviest decile is well above $200/mo.
  • Unit economics: interchange 1.5–2.0% + interest income (Greenlight pattern, 5.99–35.99% APR on the longer-financed cohort) + subscription tier ($5.99–$24.98/mo) + data layer.
  • Portfolio fit: strong-but-unclaimed. SoFi covers student refinancing post-graduation; Greenlight covers under-18; the 18–28 AI-native builder cohort is uncovered. The largest unclaimed position in the thesis.
  • Status: Karat Financial covers creators. Cluely is the closest adjacent (AI-study product) but is not a fintech. Affirm has BNPL but no vertical card here.
  • Recommended position: lead a Series A or pre-emptive seed in 2026. The gap I would source against directly.

Sub-thesis C · Infrastructure

The AI-spend data layer

A B2B SaaS or data marketplace that aggregates AI-spend signal across the cards above (and from API integrations with employers, schools, and AI tool providers directly), and sells it to banks underwriting credit, employers underwriting talent, and AI vendors measuring PMF.

  • Pattern: Plaid × Affinity × Harmonic × SemiAnalysis. Vertical data infrastructure that becomes the source of truth for a market the incumbents do not measure.
  • Buyer: banks (underwriting), enterprises (talent), AI vendors (PMF measurement), VCs (deal flow scoring).
  • Unit economics: SaaS subscription ($30K–$300K ARR enterprise tier) + per-query data API + benchmark reports.
  • Portfolio fit: the cleanest pattern match to the Bill.com infrastructure approach. Multi-sided buyers.
  • Status: empty. Affinity, Harmonic, and SemiAnalysis are adjacent but not AI-spend-specific.
  • Recommended position: seed bet, opportunistic. Likely emerges after cards from A and B reach scale. Watch the field.

Comparable Landscape

The thesis lives inside a defined comp pattern: identify a distinctive group of people with predictable value, build a card or financial product for them, capture the alt-data, underwrite on the dataset others cannot see.

Comp Founded Pattern Last marker Read
Cherry 2017 Elective healthcare BNPL, alt-data underwriting $1.5B+ annual loan volume (2026), $2B+ financed since launch Clearest live precedent. Two-sided economics that beat the incumbent (CareCredit) on both sides.
SoFi 2011 Student refi + cross-sell into a fintech stack Public (NASDAQ: SOFI) Student-fintech-as-wedge. Started as student refi, compounded into a bank.
Greenlight 2014 Debit card + financial-literacy overlay (under-18) $550M+ raised, $2.3B valuation (Series D, April 2021), 6M+ users Card-plus-outcome-overlay. Subscription tiers stacked on card economics. Closest model for the consumer AI card.
Karat Financial 2019 Credit card for creators, underwritten on platform earnings $26M Series A in 2021 ($11M equity + $15M debt, USV-led) Alt-data underwriting works for non-W-2 populations. Direct template for the AI-builder card.
Affirm 2012 Real-time BNPL underwriting on permissioned data Public (NASDAQ: AFRM), ~$22B market cap (June 2026) Spark + Databricks at point of sale. The real-time-data playbook.
Bill.com 2006 Vertical financial-operations infrastructure Public (NYSE: BILL), ~$3.5B market cap (down from $20B+ peak) Infrastructure-compounds pattern. Started narrow, ended as the network underneath SMB financial ops.
Mercury 2017 Banking for startups + spend management $3.5B post-money (Series C, Sequoia-led, March 2025); $500M 2024 revenue Closest current standalone benchmark for a vertical card + banking-stack startup. 10 consecutive quarters of GAAP profitability.
Ramp 2019 Corporate card + spend management $32B valuation (Lightspeed-led, November 2025), up from $7.65B in early 2024 Horizontal. AI is a fraction of their volume. The compression of corporate-card moats — and the speed of the up-round — is the timing argument for vertical entry.
Brex 2017 Corporate card + spend management Acquired by Capital One for $5.15B (early 2026), down from $12.3B peak (Series D, 2022) Horizontal-card consolidation has begun. The reason vertical timing matters: the generalists are being absorbed or revalued, not extending.

The category does not yet contain a winner. Karat, Cherry, and Greenlight are the closest pattern matches; none of them are AI-vertical.

Recommended Actions

If the thesis holds, three concrete next moves. None blocks the others.

  1. Open one consumer-card seed bet by Q4 2026. Target the AI-builder cohort (sub-thesis B). Lead a $2–5M seed at $10–25M post or pre-empt a Series A round at $40–80M. Comps for valuation: Karat's $26M Series A (2021), Greenlight's seed at ~$10M post (2014). The right partner relationship to chase: founder team with prior consumer-fintech + a co-founder embedded in the AI-builder community.
  2. Watch the corporate side; don't lead. Sub-thesis A will produce 3–5 named entrants in 2026. Wait for a traction proof point — interchange run-rate by month 12, 25+ paying corporate customers — before leading a Series A.
  3. Map the data layer (sub-thesis C) as a seed-watch portfolio. The infrastructure bet is too early for a lead position today but worth holding 3–5 seed conversations open through 2027.

Geographic sequencing: US-first for both card bets. HK / Singapore / Tokyo expansion in year 2 — the cross-border arc that MoneyForward executed cleanly.

Capital reserve thesis: if both card bets reach scale, the data layer (sub-thesis C) becomes the natural acquisition target. Hold reserves for the infrastructure consolidation in 2028–2029.

Falsifiability

This thesis is wrong if any of the following are true by mid-2027.

  1. Pilot data shows no underwriting lift. The first 200–500 AI-spend cardholders in a live pilot show no measurable difference in repayment behavior, default rate, or income progression versus a FICO-matched control after 12 months of card-on-file data. If the moat is invisible at pilot scale, the data-moat claim is dead.
  2. Incumbents extend faster than expected. Brex, Ramp, or Mercury ship AI-spend categorization + benchmark dashboards by Q2 2027, foreclosing sub-thesis A.
  3. Card economics compress. Interchange compression in the US drops blended take rate below 1.5%, killing the unit economics for sub-thesis B at the under-30 segment's AOV.
  4. AI providers refuse partnership. Fewer than 3 major AI vendors sign cashback or co-marketing deals with any vertical card entrant within 12 months. Distribution kills sub-thesis B at the gate.
  5. Consumer regulatory tightening. A material shift in US consumer-credit regulation around alt-data underwriting (CFPB, state-level) raises compliance cost above the AOV economics will support. Especially watch the student segment.

Each of these is testable on a defined timeline. If any two trigger together, the thesis should be revised before further capital is committed.