Skip to content

Idea Scorecard

Phase: 8 — Validation (final) Project: likeness Date: 2026-05-09 Confidence: Honest — calibrated to research and explicit about what's still unknown

Scorecard

Dimension Score (1-10) Rationale
Problem severity 8 Unauthorized AI use of adult creators is large, growing (8M+ deepfake incidents 2025, +900% in two years), and harms creators with no current recourse. Public sentiment is strongly negative on adult deepfake abuse (47.5% negative — highest of any deepfake topic). The problem is real, named, and getting worse. The frequency-and-intensity score is high.
Market size 6 Macro markets (adult creator subscription + adult AI generation) are sizable ($7.22B OnlyFans GMV; $1-3B adult AI generation). However, the US-first MVP SAM ceiling is ~$60M ARR, below typical Series A benchmarks. Reaching venture-scale requires expansion (video, voice, geos) — credible but unproven path. Yellow Flag.
Competitive advantage 7 Likeness occupies the only uncontested cell at the intersection of consent-first + explicit-allowed + verified-creator-licensed + strong architectural commitments. Vylit ($2.7M Seed, ex-OnlyFans CEO) is the closest direct comp but explicitly excludes explicit content. Fanvue could expand. The cell is open today; window is finite (12-18 months estimated).
Feasibility 7 ML architecture is buildable from off-the-shelf primitives (LoRA per creator, IP-Adapter / InstantID, ControlNet, off-the-shelf watermarking). Real engineering, not research. The genuinely hard part is operational compliance (2257, multi-processor BD, takedown ops at scale) — well-named in the founder brief and resourced via specialty cofounder seats.
Business model clarity 6 Revenue mix is benchmarked clearly (OnlyFans 59%-one-off vs. 41%-subscription informs design). Creator-take net of platform + processor fees is ~65-75% — workable but materially below OnlyFans (~78%). Most leveraged variable (% of fans on AI tier) is unvalidated. Yellow Flag.
Founder-market fit 5 The honest score. Founder is a senior CTO with strong product/engineering reflexes — appropriate for the platform/architecture work. Outside-in to the adult industry, with a small network of insider friends who can introduce. Creator-side trust will rest on the Creator Ops cofounder hire. Vylit's ex-OnlyFans CEO founder profile is what the cofounder hire must approximate. Yellow Flag.
Timing 8 Strong regulatory tailwinds: TAKE IT DOWN Act enacted, NO FAKES Act pending, California AB 2602 in effect, C2PA standardization accelerating, EU AI Act Article 50 enforcement August 2026. Mainstream consent infrastructure has cleared Series A; adult-side equivalents are at Seed scale with one funded competitor (Vylit). The category is forming now.
Overall 6.7 Conditional proceed.

Verdict — CONDITIONAL PROCEED

Score: 6.7 / 10. Verdict: GO, conditional on three specific validations.

The research strongly supports the thesis at the macro level: the consent-first AI likeness platform is structurally aligned with regulatory direction, has a real and growing problem to solve, and occupies a competitively uncontested cell. The product architecture is buildable. The capital plan is plausible.

Three conditions must validate before deeper capital commitment beyond the pre-seed close:

  1. Creator participation must validate via structured interviews. ≥7 of 10 mid-tier verified adult creators interviewed must express genuine interest in the concierge configuration as proposed. This resolves the load-bearing unknown that no amount of research can resolve.

  2. At least two adult-friendly payment processors must indicate approval pre-launch. Single-processor dependency is a kill-the-company risk per Civitai precedent. Multi-processor redundancy operational at launch is non-negotiable.

  3. The Creator Ops cofounder hire must close within 90 days of funding close. This is the determinant of creator-acquisition velocity AND competitive parity vs. Vylit's founder-market fit. The founder cannot bridge this gap personally.

If any of these fails, the appropriate response is pivot or stop, not push-through. The kill criteria in 06-validation/validation-playbook.md specify the precise thresholds.

What would move this to GREEN (8+)?

  • Strong creator-discovery validation (8-9 of 10 saying yes)
  • Two processor agreements signed in principle pre-launch
  • Creator Ops cofounder hired within 60 days at signed terms
  • Concierge cohort with 5+ committed creators in the first cohort

What would move this to RED (≤4)?

  • Creator-discovery comes back ≤3 of 10 saying yes — the macro thesis works but the micro implementation doesn't with this specific creator base
  • Payment processors decline based on category — the entire category is not viable through legitimate channels
  • A funded direct competitor with explicit-content scope and architectural commitments enters the market within 6 months — competitive cell becomes contested

What we know vs. what we're guessing

We know (well-supported)

  • The regulatory direction strongly favors this design.
  • The macro market is large.
  • No funded direct competitor in the explicit-allowed cell exists today.
  • Fans pay for AI on creator platforms (Fanvue 15% AI revenue is concrete).
  • The architecture is buildable.

We're guessing (inferred-not-validated)

  • Creators with existing audiences will participate at the proposed economics.
  • The proposed pricing tier structure resonates with both creators and fans.
  • The fan adoption rate of AI tiers will reach 15%+ within 90 days.
  • Adult-friendly processors will approve a consent-posture pitch within 12 weeks.
  • The Creator Ops cofounder profile exists at cofounder-modest cash with the right insider trust.

The three conditions above test the first three guesses directly. Conditions 4 and 5 are operational risks the founder should monitor explicitly during cofounder recruitment and processor BD.

Final note on honesty

The score is genuinely 6.7, not "conservatively 6.7 because we want to call it 7." The biggest single drag is founder-market fit (5) — and this is the score that improves the most with the right Creator Ops cofounder hire. A successful Creator Ops cofounder hire moves the overall score to 7.5+ and the verdict toward strong GO. That hire is the highest-leverage single action between today and concierge launch.

Strategic Connections

  • The "conditional proceed" verdict directly maps to the gating experiments in 06-validation/validation-playbook.md.
  • The founder-market-fit score is supported by the founder Q&A in 00-intake/brief.md and the competitive analysis in 01-discovery/competitor-landscape.md.
  • The market size score reflects the SAM ceiling analysis in 01-discovery/market-analysis.md.

Flags

Red Flags: - The 6.7 score is real, not optimistic. A founder reading this should not interpret it as "basically 8."

Yellow Flags: - The score is heavily contingent on cofounder hires that haven't happened. If Creator Ops or CEO recruiting drags, the score effectively drops as time passes without those hires materializing.

Sources

  • All research in 01-discovery/
  • Strategy in 02-strategy/lean-canvas.md
  • Financial in 05-financial/revenue-model.md
  • Validation in 06-validation/validation-playbook.md
  • Founder Q&A this session