Asi puedes editar videos con IA (sin saber de edicion)

Asi puedes editar videos con IA (sin saber de edicion)

This is how you can edit videos with AI (without knowing about editing)

Title history
  • Asi puedes editar videos con IA (sin saber de edicion) — observed on 20261114-0303 UTC (YouTube)
🎙 EDteam 👥 1.0M 📅 November 14, 2026 ⏱ 11 min 👁 658 📄 tutorial 🧭 2026-11-14
Available in: English (current) Français

Keywords

MCPDaVinci Resolveskillsediting workflowHDR color space

Summary

The video provides a live, unscripted demonstration of editing video using artificial intelligence tools, specifically connecting ChatGPT (via the Codex interface) with DaVinci Resolve through the Model Context Protocol (MCP). The host explicitly rejects the common ‘prompt and fight’ approach, advocating instead for the creation and iterative refinement of reusable ‘skills’ that encode an entire production workflow. He records multiple raw clips, places them in a local folder, and instructs the AI to perform the edit. DaVinci Resolve, driven by MCP, automatically imports the clips, schedules them on a timeline, removes silences, transcribes the content, and even detects errors. A crucial distinction is drawn between MCP (direct tool communication) and Computer Use (screen-simulation ), with a strong warning against the latter due to inefficiency. The demonstration also exposes a color space issue (HDR) that was not in the skill, leading to a correction and an update to the skill. The host concludes by arguing that ‘good editors’ will adapt to this paradigm by maintaining and updating their AI skills, while ‘bad editors’ will be replaced. The presentation is interspersed with promotional segments for EDteam’s training courses and a Black Friday sale.

191 words

Critical Evaluation

Value of the Information & Strength of the Argument

The primary value lies in the concrete, step-by-step demonstration of a modern AI-integrated editing pipeline, which is immediately actionable. The explanation of MCP versus Computer Use offers a practical technical insight that can save users significant time and resources. The argument that building a structured ‘skill’ (workflow) is more effective than merely sending isolated prompts is logically presented and well-illustrated through the live demo, which fails initially on the HDR issue and is then corrected. However, the overall argumentation has a strong subjective and provocative tone. Claims such as ‘bad editors will lose their jobs’ are statements of opinion rather than data-driven predictions. The reasoning is based on the host’s personal experience at his own company and does not present a balanced view of the broader implications for the industry. The proposal to monitor and update AI skills to overcome contextual limitations is sound, but the video does not explore potential failure modes or ethical considerations beyond the immediate workflow.

Scientific Rigor, Source Quality, Title Accuracy

The scientific rigor is moderate. The video is a practical tutorial rather than a research-based analysis, and it does not cite external studies or literature to support its claims. The main ‘sources’ are the tools themselves and the platform’s own course links. The title accurately matches the content of the video, which is a direct demonstration. The presence of a sponsored segment (approximately 2 minutes, starting at 00:00 and again around 06:30) promotes EDteam’s enterprise AI training and Black Friday discounts, but this does not affect the technical accuracy of the demonstrated workflow. The workflow relies on proprietary tools (ChatGPT, DaVinci Resolve, Notion) without discussing alternatives or open-source options, which limits its impartiality as a scientific review. No user comments were provided for analysis, so the audience reception cannot be assessed.

306 words

Title / Content Match

The title accurately reflects the core content: a live demonstration of AI-driven video editing without manual intervention. The hype tone is present but does not deviate from the promised scope.

Quality & Reliability

6/10

The demonstration is practical and reproducible, showing a coherent workflow with established tools (ChatGPT, DaVinci Resolve, MCP). However, the content is heavily opinionated and promotional, with strong subjective claims about the future of the profession that lack empirical support. The live demo partially validates the claims but introduces variables (e.g., HDR issues) that are handled ad hoc.

Key Moments

Cited Sources

External References

Contribution & Novelties

The video contributes a concrete, replicable workflow for autonomating video editing using Large Language Models via MCP, a concept not widely covered in mainstream tutorials as of late 2026. The emphasis on building a persistent, evolving ‘skill’ (state-based workflow instructions) rather than using static prompts represents a notable method for professional AI-assisted editing. It also clarifies the technical superiority of MCP over Computer Use for tool integration, a practical insight for content creators.

Pour aller plus loin :

137 words

Radar Profile

The radar profile shows high scores in information quantity and technique level, reflecting a detailed and hands-on demonstration, but a slightly lower information quality and a significantly lower global reliability score due to subjectively presented arguments and a promotional undercurrent. The user receives substantial practical value but should remain critical of the strong opinions.

Reliability 5/10