J’ai enterré Google trop vite — Voici le nouveau Gemini !

J’ai enterré Google trop vite — Voici le nouveau Gemini !

by Ludo Salenne

Humanities, Social Sciences & Thought AI — Models & Research UYQArtificial intelligence (AI)UYQFGenerative artificial intelligence
🎙 Ludo Salenne 👥 267K 📅 December 26, 2024 ⏱ 33 min 👁 41K 📄 expert opinion 🧭 2026-08-21
Available in: English (current) Français

Keywords

Gemini 2.0Deep ResearchAI OverviewChatGPT Searchcomparison

Summary

In this video, Ludo Salenne revisits his previous claim that Google is no longer the best search engine, prompted by the release of new Google AI features. He tests Gemini Advanced (1.5 Pro), the AI Overview in Google Search, Gemini Memory, and the new Gemini 2.0 Experimental Advanced model. He compares these with ChatGPT Search and ChatGPT’s o1 Pro mode. He demonstrates that Gemini’s AI Overview provides instant, well-sourced answers for simple queries, but lacks personalization compared to ChatGPT Search, which uses memory and custom instructions. He highlights Gemini’s Deep Research feature, which generates comprehensive reports by scanning multiple sources, but notes it is only available in English and not in France/EU. He also tests Gemini 2.0, which he finds to be a strong model but not clearly superior to ChatGPT. He concludes that while Google has made significant strides, ChatGPT Search remains more accessible and personalized for his use case, but he acknowledges that Google’s AI features are impressive and may change the competitive landscape.

166 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides valuable hands-on demonstrations of Google’s AI features, showing real-world performance and limitations. The creator’s argumentation is based on personal experience and direct comparisons, which is useful for viewers considering these tools. However, the evaluation is subjective and lacks controlled testing or quantitative metrics. The creator acknowledges the limitations of his tests, such as language barriers and regional availability, which adds credibility. The argument that ChatGPT Search is more personalized due to memory and custom instructions is well-supported by demonstrations. The comparison with LMArena rankings provides an external reference, but the creator notes the absence of o1 Pro mode from the leaderboard, showing critical thinking.

Scientific Rigor, Source Quality, Title Accuracy

The video does not cite formal scientific sources but relies on direct testing and the LMArena leaderboard. The creator mentions his own previous videos and provides links to them, which serve as references for his claims. The title accurately reflects the content, as the creator indeed revisits his previous stance and tests the new Gemini features. The video is a subjective opinion piece rather than a rigorous scientific analysis, but the creator is transparent about his methods and limitations. The description includes links to his own tutorials and resources, which are relevant but not external authoritative sources.

219 words

Title / Content Match

The title accurately reflects the content: the creator revisits his previous criticism of Google and tests the new Gemini models and features, concluding that he may have been too hasty.

Quality & Reliability

6/10

The video is a subjective hands-on test of Google's AI features, based on personal experience and anecdotal evidence. It lacks rigorous methodology, but the creator is transparent about limitations and provides practical demonstrations. The claims about model rankings are based on a community leaderboard (LMArena) which is a legitimate but subjective source.

Chapters

Cited Sources

Concurring Sources

  • LMArena Leaderboard — The creator references this leaderboard to support the claim that Gemini 2.0 is currently the best AI model.

Contribution & Novelties

The video provides a practical, hands-on comparison of Google’s new AI features (Gemini 2.0, Deep Research, AI Overview) against ChatGPT, highlighting real-world usability and regional limitations. It offers insights into the current state of AI-powered search and the competitive dynamics between Google and OpenAI.

Pour aller plus loin :

  • LMArena Leaderboard — Community-based AI model ranking used in the video.
  • Google Gemini — Official page for Google’s AI assistant.
  • ChatGPT — OpenAI’s AI assistant, the main competitor discussed.

78 words

Radar Profile

The radar profile shows moderate scores across all dimensions, with a slight emphasis on quantity of information and lower technical depth. This reflects a practical, opinion-based review rather than a deep technical analysis.

Reliability 5/10