Je teste vos GPTs sur CHATGPT (vous êtes des CRACKS !)

Je teste vos GPTs sur CHATGPT (vous êtes des CRACKS !)

by Ludo Salenne

Applied Sciences & Engineering AI — News & Applications UYComputer scienceUYQArtificial intelligence (AI)
🎙 Ludo Salenne 👥 267K 📅 November 19, 2023 ⏱ 46 min 👁 20K 📄 news review 🧭 2026-08-21
Available in: English (current) Français

Keywords

GPTsChatGPTtestpersonnalisationIA

Summary

In this video, Ludo Salenne tests a selection of custom GPTs created by his subscribers, showcasing the diversity of applications and the current capabilities of the GPT builder feature. He begins with Slide Assistant, a GPT that generates PowerPoint presentations via VBA scripts, noting its usefulness but also the need for manual adjustments. Next, he explores Narrative Nexus, a creative writing assistant, and tests its security by attempting to extract its training data, revealing that GPTs are not fully secure. He then evaluates CTS, a trading coach, which provides educational content but hesitates to give direct market analysis, though it eventually analyzes a Bitcoin chart when prompted. Master Prompt, based on his own ‘Promptor’ technique, is praised for its effectiveness in optimizing prompts. He also tests a pedagogical GPT, Geppetto, a parental coach, a story weaver, and a legal assistant, each with varying results. Throughout, he encounters technical issues (‘vidéo de la poisse’) and discusses the importance of GPT configuration and security. The video concludes with a call for an episode 2 and encourages viewers to share their own GPTs.

180 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides valuable, hands-on insights into the practical use of custom GPTs, highlighting both their potential and current limitations. The creator’s argumentation is based on direct testing and personal experience, making the information relatable and actionable. He demonstrates critical thinking by probing the security of a GPT and noting the need for better automation in some cases. The feedback is constructive, offering specific suggestions for improvement to each creator. However, the evaluation is subjective and lacks quantitative metrics or comparison with alternative tools.

Scientific Rigor, Source Quality, Title Accuracy

The video is a practical review, not a scientific study, so the rigor is appropriate for its purpose. The creator clearly states his methodology (testing each GPT with specific prompts) and provides links to all the GPTs in the description, allowing viewers to verify and try them. The title accurately reflects the content, and the video’s structure with chapters helps navigation. The creator also mentions his own training course, which is a commercial element but clearly disclosed. Overall, the sources are transparent and the title-content alignment is good.

187 words

Title / Content Match

The title accurately reflects the content: the creator tests user-submitted GPTs, showing both successes and failures, with a playful tone.

Quality & Reliability

7/10

The video is a practical review of custom GPTs, based on direct testing and personal experience. The creator provides transparent feedback and mentions limitations (e.g., message caps, security issues). However, the evaluation is subjective and lacks external verification or systematic methodology.

Chapters

Cited Sources

Concurring Sources

Contribution & Novelties

The video provides a unique, community-driven perspective on the early days of custom GPTs, showcasing real-world applications and user creativity. It offers practical insights into the strengths and weaknesses of the GPT builder, including security vulnerabilities and the importance of prompt engineering. The creator’s interactive approach, testing GPTs from his audience, adds a novel and engaging element to the content.

Pour aller plus loin :

98 words

Radar Profile

The radar profile shows high scores in information quantity and quality, reflecting the video's rich content and practical insights. The technical level is moderate, suitable for a general audience, while the global reliability is good, though limited by the subjective nature of the review.

Reliability 7/10

💬 Très positif. Sur les 30 commentaires analysés, l'enthousiasme est unanime : les spectateurs saluent la qualité de la vidéo, remercient pour les découvertes, et réclament un épisode 2, malgré les aléas techniques rencontrés.