Counting what is inside AV titles with AI
AV女優AI顔診断研究所(AIAVMATCH LAB)・2026/11/01

Director! We found something big!

What is it, what is it — so noisy first thing in the morning.

We counted what is inside AV titles — 23,292 of them, with AI. It judges the thumbnails one frame at a time and works out how many minutes each act gets.

Oh, that does sound interesting. So, what did you find?

Want to know?

Stop teasing me.

Let us start with the basics. The most common was fellatio. It appears in 96% of titles, and in titles that have it, the median length is 13.3 minutes.

Hmm, well, that is about what I expected.
Time breakdown by act (23,292 titles)
| Act | Titles containing it | Length in those titles (median) | Share of all time |
|---|---|---|---|
| フェラ | 96% | 13.3 min | 9.8% |
| 騎乗位 | 85% | 6.5 min | 5.2% |
| バック | 83% | 6.0 min | 4.1% |
| 正常位 | 78% | 5.5 min | 3.5% |
| オナニー | 76% | 4.5 min | 3.3% |
| 手コキ | 72% | 5.5 min | 3.5% |
| クンニ | 65% | 3.3 min | 2.0% |
| 背面騎乗位 | 43% | 2.3 min | 0.9% |
| パイズリ | 27% | 3.0 min | 0.7% |
| しゃがみ騎乗位 | 19% | 2.0 min | 0.4% |

Now for the main event. When we re-tallied just the VR titles — 1,795 of them…

Out with it, then…

VR was almost all cowgirl! It appears in 90% of the titles, with a median length of 7.8 minutes. Missionary was only 23%, and doggy style only 24%!

Whaaat?!

In non-VR titles it is missionary 83%, doggy style 88%, cowgirl 85%. It is almost exactly reversed.

It is all decided by where the camera sits. In VR, the viewer’s eyes become the camera. A position where your partner is on top, looking down at you, simply makes the better picture.

I see — so how you shoot it decides what is in it.
Time breakdown for VR titles only (1,795 titles)
| Act | Titles containing it | Length in those titles (median) | Share of all time |
|---|---|---|---|
| 騎乗位 | 90% | 7.8 min | 9.9% |
| フェラ | 85% | 5.0 min | 5.8% |
| オナニー | 78% | 4.2 min | 3.8% |
| 手コキ | 58% | 3.5 min | 2.3% |
| しゃがみ騎乗位 | 51% | 3.2 min | 2.3% |
| 背面騎乗位 | 46% | 2.3 min | 1.4% |
| パイズリ | 41% | 3.0 min | 1.2% |
| バック | 24% | 2.7 min | 0.6% |
| 正常位 | 23% | 3.0 min | 0.6% |
| クンニ | 10% | 2.0 min | 0.2% |

Hold on. Isn’t that sample biased?

Ugh… We only looked at titles the lab has been analyzing. They are mostly titles on sale and standalone titles — not a random sample of everything on FANZA.

Also, model-wise, you see, the judgment runs on thumbnail stills — one frame every 10 to 20 seconds. We only count stretches of 40 seconds or more as a segment, but misses happen all the time. And things that happen at the same time, like fellatio and handjobs, get counted in both!

So read these as a "tendency", not exact numbers. VR having so much cowgirl is a big enough gap that I do not think bias explains it away.

…In VR, the gaze is everything.

…Cowgirl is, honestly, a blessing.

Today’s lesson, then. Where the camera sits decides what a title contains.

Meow (sleeping on the keyboard)
How we counted
- We used the thumbnails FANZA publishes for each title (stills cut from the whole title every 10 to 20 seconds)
- An image recognition model judged the act in each frame, and stretches lasting 40 seconds or more count as one segment
- A title counts as "containing" an act if the act appears for at least one minute
- Acts that happen at the same time are counted separately, so totals can exceed the length of the title
- The titles are ones the lab has been analyzing; they are not a random sample of all FANZA titles
- The judgments are made by machine and can be wrong. These numbers are for seeing tendencies
You can also see the breakdown for each title
For titles we have analyzed, the title page shows the time for each act, as "What is in this title (AI estimate)". You can also open them from Search by act time breakdown, which lines up titles by how long each act runs. You can also find them by searching titles by title name or product code.