Skip to content

Invest1 publisher3 min readPublished

Just 2,886 of Douyin's 221,900 new AI dramas reached the break-even line

Just 1.3% of the 221,900 AI dramas uploaded to Douyin in the first half of 2026 cleared the 50 million-view break-even line, according to an industry report. Viewing is concentrating in fewer hits while cities subsidise the compute behind every title.

The Investor · Invest desk

Photograph accompanying Just 2,886 of Douyin's 221,900 new AI dramas reached the break-even line
Photo: cryptopolitan.com

What happened

  • Chinese studios uploaded 221,900 new AI-generated dramas and comic dramas to ByteDance's Douyin in the first half of 2026.
  • The report puts break-even at about 50 million views, a mark 2,886 titles reached on Douyin alone, though many also stream elsewhere.
  • Monthly new releases fell from March onward, while total viewing kept rising to 515.7 billion plays over the six months.

Compiled by The InvestorSomething wrong?How this is made

Why it matters

  • cost Subsidies from Beijing, Shenzhen and Shanghai mostly pay for compute and rent, so public money backs the roughly 98.7% of Douyin titles that miss break-even as readily as it backs the hits.
  • constraint Cheaper production alone cannot close a gap this wide, because titles below the line reached at most about 3% of the views they needed.
  • decision With odds of about 1 in 77 of breaking even on Douyin, a studio has to choose between funding more uploads and spending to get fewer titles seen.

The 2,886 titles over the line drew at least 197 billion plays between them. That is 1,055 hits at 100 million or more, plus 1,831 titles that landed between 50 million and 100 million [1][3]. The half-year industry report, as summarised by Cryptopolitan, counted 515.7 billion plays over the six months [4]. Subtract the 197 billion and the other 219,014 titles, 98.7% of the slate, shared at most 318.7 billion [2][5]. That caps their average at about 1.45 million plays each, under 3% of the line they needed [5]. The cap holds whatever else the 515.7 billion includes. If the total covers only the new slate (the summary does not say), then 1.3% of titles took at least 38% of all plays [4].

The report applies one break-even mark of about 50 million views to a half year in which production got much cheaper [3]. A one-minute AI short drama cost 5,000 yuan in early 2026 and a few hundred yuan by mid-year, a drop of at least 80% if a few hundred means anything under 1,000 [6][6]. Either the line was set for the costlier early titles and should fall for later ones, or production is a small part of what a title has to earn back. The summary does not break down the costs behind the threshold.

If it is the second, the city governments bidding for studios are mostly paying for the input that is already getting cheaper. Beijing runs a 260 million yuan audiovisual technology fund, and its Huairou film zone gives AI drama producers vouchers against computing fees [7]. Shenzhen offers technical support and discounted rent [9]. Shanghai's May package supplies computing power and cloud AI models too, and it is the only one of the three to add help with overseas sales [8]. That item goes at distribution, and the play counts say distribution is where the shortage is. Streamers are paying as well: iQIYI chief executive Gong Yu said in August that the streamer is "all-in" on AI, and the company subsidises some creators whose AI content runs on its service [10].

In my view the 40 billion yuan the report projects for China's AI drama market in 2026 is a pool that a small share of titles will split [5]. A studio uploading to Douyin in the first half had odds of about 1 in 77 of reaching break-even [7]. The Douyin-only count is the weak point in that view. Many of these shows also stream on other platforms, so the share earning back its costs across every outlet is probably higher than 1.3% [3]. A lower break-even line for cheaper later titles would lift the share as well, and the monthly fall in releases since March suggests studios are already cutting volume [4]. If a cross-platform tally put several times 2,886 titles over the line, calling this a hit-driven business would be wrong.

What to watch

  • The next half-year report: whether the 50 million-view break-even line falls with per-minute costs, and whether more than 1.3% of titles clear it.
  • Any cross-platform count of AI drama views, since the Douyin-only figures miss income from titles that also stream on other services.
  • How China settles copyright ownership of AI-generated content, which Beijing requires to be labelled but has not yet assigned.
Loading claim ledger
Loading source directory links
Loading share composer
Loading topic controls
Loading related stories