AI把微短剧的制作成本打下来了，监管把合规门槛提上去了。
8月18日消息，广电总局网络视听司发布最新《管理提示》，进一步明确真人微短剧、AI微短剧、互动微短剧的分类分层实操标准，为即将于9月1日施行的《微短剧发展管理办法》铺路。
核心数字很简单：真人微短剧投资300万以上才算一类（重点管理），AI微短剧投资80万以上就按重点管理。中间差了220万。
为什么AI的门槛反而低？
表面上看，AI微短剧80万的「重点线」比真人剧的300万低很多，似乎监管更严。但这个数字恰恰反映了一个行业现实：AI制作成本就是真人的几分之一。
传统真人微短剧，要找演员、搭场地、拍片子、做后期，几十万元预算可能连开机都难。AI微短剧呢？一个人、一台电脑、一个模型，几万块钱就能产出几十集。2026年一季度，全行业上线微短剧约12.8万部，其中AI短剧占比超95%——这个比例本身就说明了问题。
所以80万的AI重点线，对应的其实是和300万真人剧相当的「项目体量等级」。它不是针对AI的歧视，而是根据实际生产成本做出的校准。
当然也有独立于投资额的硬红线：只要涉及政治、军事、外交、国家安全、统战、民族、宗教、司法、公安等特殊题材，无论投资多少，都按最高等级管理。AI生成内容在重大议题上的风险，监管看得很清楚。
从野蛮生长到精耕细作
监管收紧的背后，是行业乱象的持续发酵。
2026年4月至今，全网AI短剧下架总量已超7万部，新剧过审率骤降至30%以下。盗脸盗声、剧本洗稿、低俗猎奇、价值观跑偏……低门槛带来的不仅是产量爆炸，也是泥沙俱下。
新规之后，行业会加速洗牌。大额精品项目的合规成本抬升，倒逼资本向头部制作机构集中。靠批量生产同质化内容赚快钱的小作坊，日子会越来越难——要么升级做精品，要么转型做海外，要么直接出局。
但监管不是只有压力。合规体系的建立，其实也在为优质内容铺路。当劣币不再能轻易驱逐良币，真正有创作能力的团队才有机会脱颖而出。
对于AI内容产业来说，这是一个标志性的时刻：它不再是「玩具」，而是被正式纳入了主流内容监管体系。被监管，意味着被承认；有门槛，才有可能产生真正的价值。
明天见。
AI drove microdrama production costs down. Regulation is driving the compliance bar up.
On August 18, the NRTA's Network Audio-Visual Department issued updated guidelines clarifying classification standards for live-action microdramas, AI microdramas, and interactive microdramas — paving the way for the Microdrama Development Management Act taking effect September 1st.
The core numbers are simple: live-action microdramas with investment above 3 million RMB get Tier-1 (key) management. AI microdramas hit that same tier at 800,000 RMB. A gap of 2.2 million.
Why Is the Threshold Lower for AI?
On the surface, the 800K 'key line' for AI microdramas looks much stricter than the 3M for live-action. But the number actually reflects an industry reality: AI production costs a fraction of what live-action does.
Traditional microdramas need actors, sets, shoots, post-production — a budget of a few hundred thousand might not even get you started. An AI microdrama? One person, one computer, one model — you can produce dozens of episodes on a few tens of thousands of RMB. In Q1 2026, the industry saw roughly 128,000 new microdramas, with AI productions accounting for over 95%. That ratio tells you everything.
So the 800K threshold for AI key-level management actually corresponds to roughly the same 'project size tier' as 3M for live-action. It's not discrimination against AI — it's a calibration based on actual production economics.
There are also hard red lines independent of budget: anything involving political, military, diplomatic, national security, ethnic, religious, judicial, or public security subject matter gets the strictest review regardless of investment level. Regulators have no illusions about the risks of AI-generated content on sensitive topics.
From Wild Growth to Quality Focus
The tightening follows a wave of industry problems.
Since April 2026, over 70,000 AI short dramas have been taken down nationwide, and the pass rate for new shows has plummeted below 30%. Stolen faces and voices, script laundering, vulgar sensationalism, skewed values — low barriers brought an explosion of quantity but also a lot of garbage.
After the new rules, the industry will accelerate its shakeout. Compliance costs for big-budget premium projects are rising, pushing capital toward established production houses. Small workshops churning out homogeneous content will find life harder — either upgrade to quality content, pivot to overseas markets, or exit.
But regulation isn't just pressure. Building a compliance system actually paves the way for quality content. When bad money can no longer easily drive out good, genuinely creative teams get a chance to stand out.
For the AI content industry, this is a landmark moment: it's no longer a 'toy' — it's been formally brought into the mainstream content regulatory system. Being regulated means being recognized. With thresholds comes real value.
See you tomorrow.
