AIDrama98% of AI Dramas Lose Money: Why Most Creators Fail
- 98.7% of AI short dramas released in H1 2026 failed to recover production costs (DataEye-ADX, August 2026)
- 221,900 AI titles launched on Douyin alone in six months — one new drama roughly every 36 seconds
- Production costs dropped 80–85% vs. live-action, but traffic acquisition eats ~70% of total spending
- Revenue per 10,000 views collapsed from 30–100 yuan to 5–10 yuan year over year
- China's NRTA Order No. 16 took effect September 1, 2026, requiring AI-content labeling and platform accountability
DramaBoxEditor's pick — biggest short-drama library, lowest price to go unlimited
9.8Excellent
- Thousands of series, roughly 200 new titles every month
- Free episodes daily, plus ad-unlocks — no card needed to start
- Unlimited plans from about $5.99 a week, the lowest entry price we found
- Runs on Android, iOS and in a desktop browser
Affiliate link. We may earn a commission if you install through it, at no extra cost to you — it never changes our scores or placements. How we test · Disclaimer
The 98.7% failure rate, explained
About 98.7% of AI-generated short dramas released in the first half of 2026 failed to recover their production costs, according to DataEye-ADX research published in August 2026. That translates to roughly one commercially viable title for every 77 released. The benchmark used by DataEye is 50 million views — the approximate threshold at which ad revenue and in-app purchases cover a mid-range AI production budget plus its traffic-acquisition spend.
The number sounds dramatic, but it is consistent with adjacent creative industries. Mobile games, for instance, see about 95% of titles fail to recoup development costs. What makes the AI drama situation unusual is the speed at which the market saturated: monthly new releases on Douyin alone went from 655 a year earlier to over 74,000 by mid-2026 — a hundredfold increase driven by AI tools that cut production timelines from weeks to days.
This article breaks down where the money actually goes, why cheap production costs do not translate to profit, who is making money, and what the September 2026 regulations mean for creators and viewers.
How many AI dramas launched in 2026?
Roughly 430,000 micro short dramas launched on Chinese platforms in the first eight months of 2026, about 13 times the prior full-year total, according to industry data compiled by BigGo Finance. AI-generated content accounted for over 90% of that volume. On Douyin specifically, 221,900 AI-native titles appeared in H1 2026 (DataEye-ADX, August 2026). By Q2, 239,000 micro short dramas launched across platforms, of which 153,000 (64%) were AI-generated and the remainder live-action (China Net Audio-Visual Association data, August 2026).
To put the pace in context: Douyin was adding a new AI short drama roughly every 36 seconds by mid-year. In all of 2025, about 33,000 live-action micro short dramas launched nationwide. AI tools compressed the entire previous year's output into roughly four days of new uploads.
What does it cost to produce an AI drama?
A high-quality AI short drama costs 800 to 1,500 yuan (about $110 to $210) per finished minute, according to industry data reported by 36Kr in August 2026. A complete 100-minute series runs roughly 100,000 yuan ($14,000). That is one-fifth to one-seventh the cost of a comparable live-action micro drama, which starts at 400,000 to 500,000 yuan ($55,000 to $70,000) for similar runtime.
The raw AI generation cost is even lower. ByteDance's Seedance 2.0 charges about 1 yuan per second of generated footage at 720p, or roughly 6,000 yuan ($830) for a single raw pass of a 100-minute series. The jump from 6,000 to 100,000 yuan comes from quality work: creators who produce competitive content run 3 to 5 generation passes per shot, with some high-quality productions running 20 to 30 rounds per shot. Script development adds about 10,000 yuan. Post-production, sound design and quality checks account for the remainder.
The production timeline has compressed equally: 7 to 15 days for AI versus 25 to 40 days for live-action. One documented example — the drama Zombie Scavenger by creator Liu Ziyu — cost about 3,000 yuan in computing power and took 10 days. At the high end, The Laid-off Girl (73 minutes, 11 episodes) required 30 days and over 10,000 yuan per episode with 20–30 generation rounds per shot.
Seedance generation pricing
Seedance 2.0 charges about 1 yuan per second at 720p resolution. The newer Seedance 2.5 roughly doubled the rate to about 2.1 yuan per second at 720p (data as of August 2026). For a typical 2-minute episode at 720p with three generation passes, the raw footage cost is roughly 360–720 yuan ($50–$100). Higher resolution and more passes multiply the cost linearly.
AI vs. live-action: cost comparison table
| Factor | AI drama | Live-action drama | Difference |
|---|---|---|---|
| Cost per minute | 800–1,500 yuan ($110–$210) | 4,000–5,000 yuan ($550–$690) | AI is 70–85% cheaper |
| Full series (100 min) | ~100,000 yuan ($14,000) | 400,000–500,000+ yuan ($55,000–$70,000+) | AI saves $41,000–$56,000 |
| Production time | 7–15 days | 25–40 days | AI is 2–4× faster |
| Crew size | 1–5 people | 20–50+ people | AI needs 80–95% fewer people |
| Raw footage cost (100 min) | ~6,000 yuan ($830) at Seedance 720p | N/A (camera + crew) | — |
| Script cost | ~10,000 yuan | ~10,000–30,000 yuan | Comparable |
| Quality ceiling | Approaching live-action (with 20–30 passes) | Professional broadcast quality | Gap narrowing |
The cost advantage is clear, but it is also what creates the problem. Cheap production lowers the barrier to entry, which floods the market with content, which drives down the revenue each title can earn. The following sections explain why that math does not work for 98.7% of creators.
Why do AI dramas fail despite low production costs?
Production cost is only a fraction of the total investment. The primary expense for any commercial short drama is traffic acquisition — paid advertising to put the title in front of viewers. According to industry practitioners cited by 36Kr (August 2026), traffic placement accounts for roughly 70% of the total cost chain for AI quasi-live-action dramas. A title that cost 100,000 yuan to produce may need 300,000 to 500,000 yuan in ad spend to generate enough views to break even.
Three forces are working against creators simultaneously:
1. Catastrophic oversupply
With 430,000 titles launched in eight months, each new drama competes against an ocean of content. AI tools made production cheap, but every other creator used the same tools. The result: 64% of AI titles in Q1 2026 attracted fewer than 1 million views, according to BigGo Finance. On a platform where 50 million views is the break-even line, most titles never come close.
2. Traffic costs are rising, not falling
DataEye's Lin Qiwen reported that traffic (ad placement) costs rose over 100% year on year while revenue fell over 50%. More titles chasing the same viewer pool pushes ad auction prices up. The ROI on ad spend for AI dramas was only 1.03 to 1.07 by mid-2026 — meaning that for every yuan spent on ads, creators earned just 3 to 7 fen in profit, before counting production costs.
3. Revenue per view is collapsing
Revenue per 10,000 views dropped from 30–100 yuan a year ago to 5–10 yuan by mid-2026 — a decline of over 90%. This is the combined effect of oversupply (more content diluting ad revenue) and platform algorithm changes that prioritize user engagement over creator monetization. One practitioner documented uploading 11 AI comic dramas (16 episodes each) and earning a total of 9.6 yuan — less than $1.50 for weeks of work.
The traffic trap: 70% goes to ads
Traffic acquisition — the cost of paid advertising to get viewers to watch your title — is the single largest expense in the AI drama business. For a hit drama that generates 100 million yuan in recharges (in-app purchases), ad spend typically runs 60 to 80 million yuan, leaving only a few million yuan in profit after production, platform fees and operating costs are deducted (BigGo Finance, October 2026).
For smaller creators, the math is worse. The ad ROI for AI dramas sits between 1.03 and 1.07, meaning that every 1,000 yuan of ad spend returns only 1,030 to 1,070 yuan in viewer revenue — before production costs. If the title's production cost was 100,000 yuan, the creator needs approximately 3 to 5 million yuan in ad-driven revenue just to break even. Getting that level of return from a cold-start AI drama is statistically unlikely when the hit rate is 0.47%.
The business model worked when two conditions held: there were fewer titles competing for viewer attention, and revenue per view was higher. Both have reversed. The AI production boom created the oversupply, and the oversupply killed the per-view economics.
Revenue per view is collapsing
Revenue per 10,000 views has fallen from 30–100 yuan a year ago to just 5–10 yuan in mid-2026 (DataEye, via 36Kr, August 2026). That is a decline of up to 93%. For creators, this means a title needs roughly 5 to 10 times as many views today to earn what it would have earned in 2025 — while competing against 100 times as many titles.
| Period | Revenue per 10K views | Views needed for ¥100K break-even | Hit-rate context |
|---|---|---|---|
| Mid-2025 | 30–100 yuan | 10M–33M views | Achievable for a decent title |
| Mid-2026 | 5–10 yuan | 100M–200M views | Only 0.47% of titles reach 100M |
| Year-over-year change | −83% to −93% | 3–20× harder | Break-even moved from plausible to rare |
The case study that captures the new reality best: one practitioner uploaded 11 AI comic dramas with 16 episodes each (176 episodes total) and earned a total of 9.6 yuan — about $1.33. That is not a rounding error; it is what happens when low-quality content enters a market where per-view revenue has collapsed and the algorithm does not surface it.
What is the actual hit rate?
On Douyin in H1 2026, 1,055 AI drama titles exceeded 100 million views out of 221,900 released — a hit rate of 0.47%, or about 1 in 213 titles (DataEye-ADX). About 12,000 titles (5.8%) surpassed 10 million views. AI comic dramas specifically had a hit rate below 0.1%, making photorealistic AI the significantly more viable format.
The Q1 2026 hit rate was even lower at 0.16%, with 64% of AI titles drawing fewer than 1 million views. The improvement by mid-year likely reflects better AI tools (particularly Seedance 2.0's photorealistic capabilities) rather than improved market conditions — the creators who survived long enough to adopt better tools started winning a larger share of a stagnant viewer pool.
The quality gap: throwaway vs. competitive
Quality is diverging sharply in the AI drama market. At the low end, mass-produced AI comic dramas (non-photorealistic, often with minimal post-production) attract fewer than 1 million views 64% of the time and have a hit rate below 0.1%. At the high end, photorealistic AI dramas using tools like Seedance 2.0 with extensive post-processing are approaching live-action visual quality and occasionally reaching 2 billion or more views.
The market share shift tells the story clearly. Photorealistic AI dramas grew from 12% of new AI releases in early 2026 to over 70% by June, according to DataEye. Their viewing share was about 17% in January, roughly 80% in April and May, and 68% in June. 3D animated dramas also rose from 6.7% of viewing share in March to 29.5% in June, with one leading title reportedly reaching 2.127 billion views.
The difference between throwaway and competitive AI dramas comes down to investment per shot. A minimum-viable AI drama runs each shot once through Seedance and publishes the raw output. A competitive production runs 20 to 30 generation passes per shot, selects the best frames, adds professional sound design and post-production, and invests in an original script rather than recycling a template. The cost difference: roughly 3,000 yuan for the first approach versus 100,000+ yuan for the second.
Who is actually making money?
Money in the AI drama ecosystem concentrates at three layers: the platforms that distribute content, the tools that generate it, and a narrow slice of top creators.
Platforms
Distribution platforms capture the largest share of value. Hongguo Free Short Drama reached over 300 million monthly active users and 168 million daily active users by July 2026 (up 107% year-over-year). Douyin pledged 500 million yuan to support live-action micro-dramas. The platforms earn from advertising, platform fees on in-app purchases, and the user engagement that short dramas drive.
AI tool providers
ByteDance's Seedance is effectively the picks-and-shovels play of the AI drama gold rush. At 1 yuan per second (Seedance 2.0) to 2.1 yuan per second (Seedance 2.5), the tool earns revenue regardless of whether the drama succeeds. Other tool providers — Kling, Pika, Runway — operate similarly. When 430,000 titles launch in eight months, the tool providers collect from all of them.
Top 1.3% of creators
The small fraction of creators who do break even tend to share common traits: original scripts rather than recycled templates, significant investment in post-production (20–30 generation rounds per shot), early adoption of the latest AI model versions, and often a track record from the live-action micro-drama era. They also tend to produce photorealistic content rather than comic-style dramas.
| Layer | Example | Revenue model | Profitable? |
|---|---|---|---|
| Distribution platform | Hongguo, Douyin, DramaBox, ReelShort | Platform fees, ads, IAP commission | Mostly yes (thin margins) |
| AI tool provider | ByteDance Seedance, Kling, Runway | Per-second or per-generation fees | Yes — usage-based, scales with volume |
| Top creators (1.3%) | Studios with 100M+ view hits | Ad revenue, IAP revenue share | Yes, but margins thin after traffic costs |
| Mid-tier creators | Semi-professional teams | Ad revenue, IAP share | Breakeven to slight loss |
| Low-end creators (64%+) | Solo/small teams, comic dramas | Ad revenue (if any) | No — 9.6 yuan on 176 episodes is common |
Platform economics: DramaBox, ReelShort and margins
Even the largest platforms in the Western-facing short drama market operate on razor-thin margins. Profitability at scale remains elusive for most, driven by the same traffic-cost dynamics that afflict individual creators.
DramaBox
DramaBox, operated by Dianzhong Technology (a subsidiary of Zhuhai Manyun Technology), earned roughly $195 million in H1 2026 in-app purchase revenue and had about 83.86 million monthly active users (Lollipop.im / DataEye, October 2026). Gross margins sit at about 10%, with net margins below 1% on short-drama operations. DramaBox was the only major Western-facing microdrama app publicly reported as profitable in 2024, earning roughly $10 million net on $323 million revenue.
ReelShort
Crazy Maple Studio, ReelShort's parent company, reported 5.721 billion yuan (~$790 million) in 2025 revenue but a net loss of 85.84 million yuan (~$12 million). Selling expenses — mainly user-acquisition advertising — consumed 57.5% of revenue, or 953 million yuan, of which 759 million yuan was direct promotional spending (BigGo Finance, October 2026). Media Partners Asia projects ReelShort will reach $1.05 billion in 2026 revenue with its first scaled profitability.
The industry pattern
Chinese Online (operator of several short-drama brands) spent 254 million yuan in selling expenses in H1 2026 alone — about 44% of revenue. Its 2025 full-year loss was 671 million yuan. Even TikTok's AI short-drama revenue sharing generated only about $2 million in Q1 2026, though that figure grew 300% quarter over quarter. The overall picture: enormous revenue flowing through the ecosystem, with almost all of it consumed by the cost of acquiring and retaining viewers.
China's NRTA Order No. 16: what changed September 1, 2026
The National Radio and Television Administration's "Measures for the Administration of the Development of Micro Short Dramas" (Order No. 16) was approved July 27, 2026, published July 31, and took effect September 1, 2026. It is the first comprehensive regulation specifically governing micro short dramas (episodes under 20 minutes) in China.
Key requirements
AI disclosure: Every episode that was generated or produced using AI must carry a visible disclosure notice. This is the most directly relevant provision for the AI drama market — it means AI-generated content can no longer be presented as live-action without labeling.
Three-tier classification: Micro short dramas are divided into three categories based on budget and production scale. Category one requires pre-production filing and a distribution license after government review. Category two requires government review at the provincial level. Category three allows distributor self-review with program-number registration. The specific budget thresholds have been debated (earlier frameworks used 1 million yuan and 300,000 yuan boundaries) but the tier structure is confirmed in the order's text.
Platform accountability: Platforms must implement editor-in-chief accountability, conduct periodic algorithm assessments, and ensure their recommendation algorithms are not designed to induce addiction or excessive spending. Platforms must also maintain credit evaluation systems for key creator accounts.
Enforcement in practice: Hongguo reportedly blocked or removed 3,522 low-quality comic dramas in one week in April 2026 and acted against more than 10,000 low-quality AI dramas by the end of April. Douyin launched an AI-drama pre-screening tool on August 3, ahead of the September 1 effective date.
Impact on creators
The regulation raises the floor for AI drama quality. Creators who were mass-producing unlabeled, low-quality AI comic dramas now face two barriers: mandatory AI labeling (which may reduce viewer engagement for clearly AI-generated content) and platform pre-screening that removes content before it reaches viewers. For high-quality creators, the regulation may actually help by reducing competition from the lowest-quality tier.
What this means for Western viewers
If you watch short dramas on apps like DramaBox, ReelShort, or ShortMax, the 98.7% failure rate affects your experience in several indirect ways.
More AI content in the catalogue. Over 74% of new Chinese short drama content in H1 2026 involved AI. Western-facing platforms source much of their library from Chinese production studios, so the share of AI-produced or AI-assisted content on apps like DramaBox is growing. Some of it is high quality (photorealistic, multi-pass production); some is not.
Platform filtering matters more. The flood of low-quality AI content makes the platform's curation more important than ever. DramaBox and ReelShort both employ editorial selection and recommendation algorithms to surface quality titles. The apps you are watching on were designed to filter out the 98.7% that failed — you are mostly seeing the 1.3% that survived.
Prices are unlikely to drop. Despite production costs falling 80–85%, subscription and coin prices on Western-facing apps have remained stable or increased. This is because production cost was never the main driver of pricing — traffic acquisition, platform margins and content licensing are. The savings from AI production are absorbed by rising user-acquisition costs, not passed to consumers.
For a detailed breakdown of what each app charges and how to watch for free, see our guide: Best free AI drama apps: how to watch without paying.
Creator survival checklist
If you are considering entering the AI drama market, the data suggests a clear set of minimum requirements to avoid being part of the 98.7%. This is not a guarantee of success — even well-prepared creators face a market with a 0.47% hit rate — but it is the baseline that separates the competitive minority from the majority that earns 9.6 yuan on 176 episodes.
| # | Requirement | Why it matters | Typical cost |
|---|---|---|---|
| 1 | Original script, not a template | Template scripts saturate within days; originality is the main differentiator | ~10,000 yuan ($1,400) |
| 2 | Photorealistic format, not comic | Comic drama hit rate is below 0.1% vs. 0.47% for photorealistic | Higher per-second generation cost |
| 3 | Multiple generation passes (10–30 per shot) | Single-pass output is immediately identifiable as AI and receives less engagement | 3–5× raw footage cost |
| 4 | Professional sound design | Sound is the weakest link in most AI dramas; viewers notice immediately | 5,000–15,000 yuan |
| 5 | Traffic budget ≥ 3× production cost | 70% of total cost is ad placement; underfunded titles never reach viewers | 300,000+ yuan for serious launch |
| 6 | Compliance with NRTA Order No. 16 | AI labeling is mandatory since Sept 1, 2026; non-compliance risks removal | Operational, no direct cost |
| 7 | Multi-platform distribution | Single-platform concentration increases risk; diversify across Douyin, Hongguo, etc. | Operational |
Pros and cons of AI drama production
| Advantages | Disadvantages |
|---|---|
| Production cost 80–85% lower than live-action | 98.7% failure rate to break even |
| 7–15 day production cycle (vs. 25–40 days) | Traffic acquisition costs eat ~70% of total budget |
| 1–5 person crew (vs. 20–50+) | Revenue per 10K views collapsed 83–93% in one year |
| Photorealistic quality approaching live-action | Hit rate of 0.47% (1 in 213 titles) |
| Rapid iteration on scripts and concepts | 430K competitors in 8 months |
| No location, casting or crew logistics | Mandatory AI disclosure (NRTA Order 16) may reduce engagement |
| Seedance 2.0 pricing as low as $0.14/second | Seedance 2.5 doubled per-second costs |
| Potential for 2B+ views on a breakout hit | Comic dramas have sub-0.1% hit rate |
FAQ
What percentage of AI short dramas lose money?
About 98.7% of AI short dramas released in H1 2026 failed to recover production costs, according to DataEye-ADX. Using a 50-million-view break-even benchmark, only about 1 in 77 titles reached profitability. The figure covers AI-native titles on Douyin; results may vary by platform.
How much does it cost to produce an AI short drama?
A high-quality AI short drama costs 800 to 1,500 yuan ($110 to $210) per finished minute. A full 100-minute series runs roughly 100,000 yuan ($14,000). Raw AI generation (Seedance 2.0 at 720p) costs about 6,000 yuan for a single pass; competitive productions run 3–30 passes per shot, multiplying the cost accordingly.
Why do AI dramas fail despite being cheap to make?
Production cost is only about 30% of total investment. Traffic acquisition (paid ads) accounts for roughly 70%. With 430,000 titles competing for viewers, ad costs have risen 100% year over year while revenue per view has fallen over 90%. The math only works for the top 1.3% of titles.
Can you still make money creating AI short dramas?
It is possible but statistically unlikely. The hit rate is 0.47% for titles exceeding 100 million views. Successful creators invest in original scripts, 20–30 generation passes per shot, professional post-production and significant traffic budgets. Minimum-investment comic dramas earn almost nothing.
What is the hit rate for AI dramas on Douyin?
0.47% in H1 2026 — meaning about 1 in 213 titles exceeded 100 million views. The Q1 rate was even lower at 0.16%. About 5.8% of titles surpassed 10 million views. Comic dramas had a hit rate below 0.1%.
Is ReelShort profitable?
Not yet as of its latest reported figures. Crazy Maple Studio reported a net loss of ~$12 million on ~$790 million revenue in 2025. Media Partners Asia projects ReelShort's first scaled profitability in 2026, with projected revenue of $1.05 billion. Selling expenses consumed 57.5% of 2025 revenue.
Is DramaBox profitable?
Barely. DramaBox had a reported net margin below 1% on short-drama operations. It earned roughly $10 million net profit on $323 million revenue in 2024 — making it the only major Western-facing microdrama app publicly reported as profitable that year. H1 2026 in-app purchase revenue was approximately $195 million.
What is NRTA Order No. 16?
China's first comprehensive regulation for micro short dramas, effective September 1, 2026. Key provisions: mandatory AI-content disclosure labels, a three-tier content classification system, platform algorithm assessments, and prohibitions on algorithms designed to induce addiction or excessive spending.
How does the AI drama failure rate affect viewers?
Indirectly. The flood of AI content means platforms must filter more aggressively. Over 10,000 low-quality AI dramas were removed from Hongguo by April 2026. Western-facing apps like DramaBox and ReelShort curate their libraries, so viewers mostly see the surviving 1.3%. The growing AI share (74% of new content) means more AI-assisted titles in your feed.
How many AI short dramas were created in 2026?
About 430,000 micro short dramas launched on Chinese platforms in the first eight months of 2026, with over 90% being AI-generated. On Douyin alone, 221,900 AI-native titles appeared in H1 2026 — roughly one new title every 36 seconds.
What separates successful AI dramas from failures?
The top 1.3% that break even typically share: original scripts (not templates), photorealistic format (not comic), 20–30 generation passes per shot, professional sound design and post-production, and a traffic budget at least 3× the production cost. The minimum-viable approach of single-pass comic dramas rarely works.
Will AI drama production costs continue to fall?
Raw generation costs may continue falling as AI models improve, but the trend is not uniform. Seedance 2.5 roughly doubled the per-second price compared to 2.0. Total production costs for competitive content are rising because quality expectations are higher. And production cost was never the barrier — traffic costs are, and those are rising.
Bottom line
The AI short drama market in 2026 is a textbook case of a technology making production easy while making success harder. AI tools cut the cost of creating a short drama by 80–85%, but the resulting flood of content — 430,000 titles in eight months — has collapsed per-view revenue by up to 93% and pushed the break-even rate down to 1.3%. The 98.7% failure rate is not a sign that AI drama is a bad format; it is a sign that a market with near-zero barriers to entry and finite viewer attention will naturally concentrate returns in a tiny minority of titles.
For viewers, the failure rate is largely invisible. The apps you use — DramaBox, ReelShort, ShortMax — are designed to surface the top fraction of content. The growing share of AI-produced series in their catalogues means you are watching increasingly sophisticated AI content, filtered by platforms that stake their user retention on quality curation. Whether AI or live-action, what reaches your screen is the survivor of a ruthless selection process.
For creators, the data is unambiguous: minimum-investment AI dramas are economically non-viable. The market rewards significant investment in original scripts, photorealistic production with extensive post-processing, and professional traffic strategies — or it rewards nothing at all. There is very little middle ground.
China's NRTA Order No. 16 may thin the field by removing the lowest-quality content and requiring AI transparency, but it does not change the fundamental economics. The failure rate is a feature of the market structure, not a regulatory gap. As long as AI tools keep the cost of production low enough for anyone to enter, the vast majority of entrants will fail. That has always been true of creative markets. AI simply compressed the cycle from years to months.
Sources
- DataEye-ADX via 36Kr — "Rampant AI-Generated Short Dramas: 220,000 New Titles Launched in Half a Year, 90% Operating at a Loss" (August 14, 2026)
- Remio.ai — "China's NRTA Tightens AI Micro-Drama Rules as 98.7% Reportedly Fail to Break Even" (August 2026)
- BigGo Finance — "Behind the explosive growth of AI short dramas: Only 1.3% of 430,000 new titles break even" (October 2026)
- Lollipop.im — "2026 AI Short Drama Trends: Scale, Landscape & 5 Shifts" (October 7, 2026)
- C21Media / NYT — "AI slashing cost of microdrama production in China to $30 per minute" (May 7, 2026)
- China Net Audio-Visual Association — H1 2026 micro short drama statistics (via 36Kr, August 2026)
- QuestMobile — Hongguo Free Short Drama MAU data (June 2026)
- Media Partners Asia via Variety — ReelShort 2026 revenue projection ($1.05B)
- National Radio and Television Administration — Measures for the Administration of the Development of Micro Short Dramas (Order No. 16), effective September 1, 2026
Affiliate disclosure. Some outbound links on this page are affiliate links, including the DramaBox links. If you install through one we may receive a commission at no additional cost to you, and it does not change the analysis we publish. Accuracy notice. Market figures, production costs and revenue estimates come from the sources cited in the text and change as the industry moves. Everything here was verified on the date stated on the page. Independence. This site is not affiliated with, endorsed by, or an official support channel for any app listed. Editorial opinion. All reviews, scores and rankings on this site represent the editorial opinion of the AI Drama Reviews team, based on publicly available information and our own testing. They are not endorsements, guarantees of quality or professional recommendations. We do not operate or control any listed application and bear no responsibility for its service, billing or content.