J’ai créé un clone de ChatGPT en 1h avec Manus AI (il déchire)

J’ai créé un clone de ChatGPT en 1h avec Manus AI (il déchire)

by Ludo Salenne

Applied Sciences & Engineering AI — Models & Research UYQArtificial intelligence (AI)UYQNNeural networks and fuzzy systems
🎙 Ludo Salenne 👥 267K 📅 March 12, 2025 ⏱ 27 min 👁 38K 📄 tutorial 🧭 2026-08-21
Available in: English (current) Français

Keywords

Manus AIChatGPT cloneAI agentOpenAI APItutorial

Summary

In this video, Ludo Salenne demonstrates how he used Manus AI, an autonomous AI agent, to create a clone of ChatGPT with additional features like image generation, audio transcription, PDF analysis, and image analysis. He starts by outlining his goal: to build a ChatGPT alternative that uses cheaper OpenAI API models instead of a monthly subscription. He provides a detailed prompt to Manus AI, specifying the desired functionalities and design inspiration from ChatGPT. The video shows Manus AI autonomously creating the project, handling errors, and eventually deploying a web application. After encountering a context length issue, Ludo restarts in a new session, providing the previous files and discussion, leading to a successful deployment. He then tests the clone: chat works with GPT-4o, image generation works with DALL-E 3, and audio transcription works with Whisper. He also demonstrates the settings page where users can input their own OpenAI API key. The video includes a hidden invitation code for Manus AI, split into four parts, for two lucky viewers. Ludo shares his experience, highlighting the potential of AI agents to build functional applications with minimal coding.

184 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides a practical, step-by-step demonstration of using Manus AI to build a functional web application. The value lies in showing the real capabilities and limitations of AI agents in a hands-on scenario. The argumentation is based on the creator’s direct experience, with visible successes and failures, which adds credibility. However, the video is largely anecdotal and lacks comparative analysis or benchmarks against other tools. The creator’s enthusiasm is evident, but the argumentation would be stronger with more objective metrics, such as cost comparison or performance evaluation.

Scientific Rigor, Source Quality, Title Accuracy

The video is a tutorial and demonstration, not a scientific study. The creator shows the process transparently, including errors and troubleshooting. Sources are limited to the creator’s own resources and OpenAI’s API documentation, which is not explicitly cited. The title accurately reflects the content, as the creator indeed builds a ChatGPT clone in about an hour. The video does not claim to be scientifically rigorous, but it provides a useful practical guide for viewers interested in AI agents.

181 words

Title / Content Match

The title accurately reflects the content: the creator builds a ChatGPT clone in about an hour using Manus AI and tests it.

Quality & Reliability

7/10

The video is a practical demonstration and tutorial, showing real-time creation of a ChatGPT clone using Manus AI. The process is transparent, with visible steps and errors, but lacks external verification and scientific rigor. The creator's enthusiasm is evident, but the content is anecdotal and not peer-reviewed.

Chapters

Cited Sources

Concurring Sources

Contribution & Novelties

The video provides a practical demonstration of using an AI agent (Manus AI) to build a functional web application with minimal human intervention. It showcases the potential of AI agents to automate software development tasks, including design, coding, and testing. The novelty lies in the hands-on experience and the specific features implemented (chat, image generation, audio transcription, PDF analysis) using OpenAI’s API models. The video also highlights the limitations, such as context length issues and the need for troubleshooting.

Pour aller plus loin :

  • AI agent — Provides background on AI agents and their capabilities.
  • OpenAI API — Official documentation for the API used in the video.
  • Vibe coding — Concept mentioned in the video, relevant to AI-assisted programming.

119 words

Radar Profile

The radar profile shows high scores in quantity and quality of information, reflecting the detailed demonstration and practical insights. The technical level is moderate, suitable for a general audience. The reliability score is moderate, as the content is anecdotal and lacks external validation.

Reliability 6/10

💬 Très positif. Sur les 30 commentaires analysés, les spectateurs expriment un enthousiasme marqué pour la démonstration, saluent la clarté du tutoriel et la performance de Manus AI, bien que certains regrettent de ne pas avoir pu utiliser le code d'invitation à temps.