7 astuces ChatGPT dont personne ne parle – Tuto 2025

7 astuces ChatGPT dont personne ne parle – Tuto 2025

by Ludo Salenne

Applied Sciences & Engineering AI — Models & Research UYQArtificial intelligence (AI)UYQFGenerative artificial intelligence
🎙 Ludo Salenne 👥 267K 📅 February 13, 2025 ⏱ 37 min 👁 68K 📄 tutorial 🧭 2026-08-21
Available in: English (current) Français

Keywords

ChatGPTastucescustom instructionsOPROo3-mini

Summary

In this tutorial, Ludo Salenne, a French YouTuber with over 300 ChatGPT tutorials, shares seven advanced tips to improve users’ proficiency with the AI. The video begins by demonstrating how to use slash commands to access tools like Canva, then emphasizes the importance of properly configuring ‘Custom Instructions’ to get personalized and high-quality responses. He shows how to repurpose these instructions to make ChatGPT adopt different personas, such as a Human Resources Director. Next, he reveals that free users can access the o3-mini model by enabling the ‘search’ and ‘reason’ features. The video then explains the OPRO (Optimization by PROmpting) method, developed by Google DeepMind, which involves using multiple AIs to iteratively refine a prompt for optimal results. He also discusses automating ChatGPT workflows, the Operator feature, and the ‘Projects’ feature for organizing conversations. Finally, he highlights the ‘Deep Research’ feature, which he finds impressive for generating comprehensive reports. Throughout the video, he promotes his paid training course ‘Automatiser ChatGPT’ and offers free resources, including a database of custom instruction personas.

171 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides a high density of practical, actionable tips that are directly applicable for users seeking to enhance their ChatGPT experience. The demonstrations are clear and well-structured, showing real-time examples of the techniques. The argumentation is largely based on the author’s personal experience and anecdotal evidence, which is persuasive but not scientifically rigorous. The OPRO method is presented as a powerful technique, but the explanation lacks depth regarding its theoretical underpinnings. The promotional segments for his paid course are integrated into the content, which may be seen as a conflict of interest, but they do not overshadow the educational value.

Scientific Rigor, Source Quality, Title Accuracy

The video does not cite external scientific sources directly, but it references the OPRO method from Google DeepMind, which is a legitimate research contribution. The description contains links to the author’s own resources and other tutorials, but no independent references. The title accurately reflects the content, and the video is well-structured with clear chapters. The information is presented as expert opinion based on extensive hands-on experience, which is valuable but not peer-reviewed. The promotional nature of some segments is transparent, but it does not undermine the core educational content.

205 words

Title / Content Match

The title accurately reflects the content: the video presents seven specific ChatGPT tips, some of which are indeed less commonly discussed. The claim of being 'secret' is somewhat exaggerated but not misleading.

Quality & Reliability

6/10

The video is a practical tutorial based on the author's extensive personal experience with ChatGPT. It provides actionable tips but lacks rigorous scientific sourcing, relying on anecdotal evidence and promotional content. The information is generally accurate and up-to-date as of early 2025, but the promotional nature and lack of external verification reduce its overall reliability.

Chapters

Cited Sources

Concurring Sources

  • OpenAI Custom Instructions Documentation — Official documentation supporting the use of Custom Instructions as described in the video.
  • OPRO Paper on arXiv — The research paper by Google DeepMind that underpins the OPRO method presented in the video.

External References

Contribution & Novelties

The video offers a practical compilation of advanced ChatGPT tips, some of which are not widely known, such as using the OPRO method with multiple AIs and leveraging free features for o3-mini. It provides a structured approach to customizing ChatGPT through ‘Custom Instructions’ and demonstrates the power of prompt engineering. The ‘Pour aller plus loin’ section suggests further exploration of these concepts.

Pour aller plus loin :

  • Optimization by PROmpting (OPRO) — The research paper by Google DeepMind introducing the OPRO method, which is central to the video’s prompt engineering technique.
  • Chain-of-Thought Prompting — A foundational paper on reasoning in LLMs, relevant to the ‘reason’ feature and o3-mini’s chain-of-thought approach.
  • Custom Instructions in ChatGPT — Official OpenAI documentation on Custom Instructions, which the video heavily relies on for personalization.

129 words

Radar Profile

The radar profile shows high scores in 'quantite_information' and 'niveau_technique', indicating a content-rich tutorial with a good level of technical detail. However, 'fiabilite_globale' and 'qualite_information' are lower, reflecting the reliance on anecdotal evidence and promotional content rather than rigorous scientific sources.

Reliability 5/10

💬 Très positif. Sur les 30 commentaires analysés, l'immense majorité exprime une forte appréciation pour la clarté et l'utilité des astuces, avec de nombreux témoignages d'achat de la formation et des remerciements pour la pédagogie de l'auteur.