PitchBook: VC funds invested $187.7M into the startups that develop deepfake technologies in 2022 and another $50M so far in 2023, up from just $1M in 2017
Margi Murphy / Bloomberg :
Context & Ripple Effects
PitchBook's deepfake tally lands squarely inside the generative-AI capital wave the related coverage documents: the $187.7M deployed into deepfake startups in 2022 arrived the same year investors pumped $1.37B+ into generative AI across 78 deals — almost matching the previous five years combined — per the 2022 generative AI funding surge.
What makes the number notable is its slope: $1M in 2017 to nearly $238M by May 2023 tracks how AI went from niche to dominant, with later PitchBook counts showing AI startups taking 36% of US deal value through Q3 2024 (Q3 US VC investment data). The deepfake line item is an early read on where capital flows when generation capabilities get cheap.
First-order effects
- VC funds now hold legible positions in deepfake toolmakers — the same PitchBook data that sizes their bets exposes those holdings to limited partners and scrutiny, turning portfolio composition itself into a reputational variable.
- With $50M already raised by May 2023, the category was pacing toward another record year, giving funded deepfake startups runway to compete on capability rather than bootstrap constraints.
Second-order effects
- As AI absorbed a rising share of overall US deal value through 2023–2024 — part of the $330B poured into roughly 26,000 AI and ML startups — generalist funds faced sharper allocation choices between frontier-model plays and application-layer niches like synthetic media.
- Capital flowing into deepfake generation raises the value of the adjacent counter-market in detection and provenance, a line item funding trackers will likely start separating if investor interest follows.
Third-order effects
- If contested niches like deepfakes keep scaling inside the broader concentration pattern — AI took a record $97B share of US startup funding in 2024 (record AI share of 2024 US funding) — the realistic policy response is subsector-targeted labeling and disclosure rules rather than blanket AI regulation, since the funding trail makes responsibility attributable to identifiable investors and startups.
The trend: Venture capital follows generative capabilities into contested applications like deepfakes faster than governance frameworks form, making funding data itself an early signal of where regulation will land.