{
  "id": 19287524,
  "title": "The hidden costs of running an AI video production business",
  "url": "https://urgent.news/2026/11/10/the-hidden-costs-of-running-an-ai-video-production-business",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-10T04:30:58.000Z",
  "source": {
    "name": "e27",
    "slug": "e27",
    "url": "https://e27.co/the-hidden-costs-of-running-an-ai-video-production-business-20261106/"
  },
  "original_language": "en",
  "account": "The AI video production business incurs significant costs, with model usage and generation tools accounting for roughly 60% of direct production expenses. The remaining portion is covered by the worker's labor, who performs tasks such as adjusting scripts, keeping production moving, and conducting quality checks. Despite the seemingly lean setup, the cost record reveals a less straightforward story.\n\nOne of the largest direct expenses is the tools and generation needed to produce the footage. Even a finished minute can hide the behind-the-scenes attempts that failed, such as inconsistent characters, changes in style, or shots that do not blend well with the preceding ones. Additional attempts incur costs and review time, making it challenging to predict expenses accurately.\n\nThe margin of the project depends on the client fee and the time spent managing delivery. Before quoting another project, it is crucial to determine which scenes may require repeated generation or specialist intervention, as running time alone provides little insight into this. Since 2025, the author has outsourced more video production work, allowing them to take on assignments that would be difficult to complete alone. However, external production absorbs about half to four-fifths of the customer fee, leaving the author responsible for their own work and keeping the project on track.\n\nCollaboration with clients and producers is a significant part of the author's role, involving clarifying the brief, explaining it to the workforce, reviewing intermediate results, and translating feedback into executable changes. A typical project undergoes around three rounds of revisions, but the impact of each revision varies greatly. Adjustments to individual images differ from changes that render the approved storyboard unusable. The author often bears the responsibility of deciding which changes can be managed and which require additional work from specialists.\n\nThe author's costs are divided into two budgets: the money paid to models, editors, and specialists, and the time invested to deliver the work. This second budget limits the number of contracts the author can manage simultaneously, forcing them to turn away less demanding projects to maintain their capacity. Repeat customers sometimes lead to lower-margin assignments, as the author believes they will require content again. However, new creative requirements or specialist effects can make each order feel like starting anew, making future work harder to value despite the presence of repeat customers.",
  "summary": "A single AI production worker produced roughly two hours of AI-assisted drama in a week. The completed project was accepted by the client. Model usage, generation and related tools accounted for roughly 60 per cent of direct production costs; my own labour made up the rest. There was no team. One AI production worker adjusted […] The post The hidden costs of running an AI video production…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}