Nano Banana, Codex, ChatGPT, on n'arrête plus les IA !

Nano Banana, Codex, ChatGPT, on n'arrête plus les IA !

Nano Banana, Codex, ChatGPT, we can't stop the AIs!

by Renaud Dékode

Humanities, Social Sciences & Thought AI — Models & Research UYQArtificial intelligence (AI)UYQFGenerative artificial intelligence
🎙 Renaud Dékode 👥 249K 📅 November 21, 2025 ⏱ 36 min 👁 41K 📄 news review 🧭 2026-09-07
Available in: English (current) Français

Keywords

Nano Banana ProGemini 3GPT-5.1 Codex MaxChatGPT GroupsAI regulation

Summary

In this episode of his weekly tech news review, Renaud Dékode covers several major AI developments from late November 2025. He starts with Google’s Nano Banana Pro, an image generation model built on Gemini 3, highlighting its multimodal capabilities, high-resolution output (up to 4K), and integration across Google’s ecosystem, including NotebookLM and advertising platforms. He notes its accessibility and pricing (around 13 cents per image), as well as its ability to edit images and generate text with perspective. Next, he discusses OpenAI’s counter-moves: GPT-5.1 Pro, a more powerful model for high-tier subscribers, and GPT-5.1 Codex Max, an agentic coding model that reportedly excels in benchmarks. He also mentions ChatGPT’s new collaborative group chats feature, which allows up to 20 people to interact with the AI in a shared conversation, and a dedicated version for teachers. The video then shifts to regulation, focusing on France’s ARCOM and its relaunch of an online hate observatory, with a new model involving partnerships with associations and public authorities. Finally, he touches on Nvidia’s record financial results and a critical report from the French Court of Auditors on the country’s AI strategy, as well as TikTok’s new digital well-being tools. The video concludes with an outro inviting viewers to share their creations.

207 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides a timely overview of recent AI product launches and regulatory developments. The author’s argumentation is largely based on official announcements and benchmark results, but he also injects personal opinions and speculative commentary. He emphasizes the competitive dynamics between Google and OpenAI, framing it as a ‘bulldozer’ vs ‘counter-attack’ narrative. While he acknowledges the impressive capabilities of these models, he also raises concerns about the societal impact, such as the difficulty in distinguishing AI-generated images and the need for regulation. The discussion of ARCOM’s new approach is informative, but the analysis remains at a surface level, without delving into the specifics of the regulatory framework or its potential effectiveness.

Scientific Rigor, Source Quality, Title Accuracy

The video does not cite specific sources or provide links to the mentioned studies or reports. The information is presented as general knowledge from the tech news cycle, which limits its verifiability. The title accurately reflects the content, as it covers multiple AI news items. The author’s tone is enthusiastic and opinionated, which may appeal to a general audience but reduces the scientific rigor. The video includes a brief mention of a sponsorship segment, which is not detailed. Overall, the content is informative but lacks the depth and sourcing expected from a scientific analysis.

220 words

Title / Content Match

The title accurately reflects the content, which covers multiple AI news items including Nano Banana, Codex, and ChatGPT updates.

Quality & Reliability

6/10

The video is a news review of recent AI developments, based on official announcements and benchmarks, but lacks in-depth verification and relies on the author's subjective interpretation.

Chapters

Cited Sources

  • Renaud Dékode's website — The author's website where discussions and additional resources are shared.

Concurring Sources

  • Google's official blog on Gemini 3 — Official announcement of Gemini 3, which underpins Nano Banana Pro.
  • OpenAI's blog on GPT-5.1 — Official announcement of GPT-5.1 and its variants.

Dissenting Sources

  • Benchmarks comparison — The video claims Codex Max outperforms other models, but independent benchmarks may vary.

Contribution & Novelties

The video synthesizes recent AI news, highlighting the rapid pace of innovation and the competitive landscape. It provides a useful overview for those following AI developments, but does not offer original analysis or deep insights.

Pour aller plus loin :

  • Gemini 3 — Overview of Google’s Gemini model family, relevant to the discussion of Nano Banana Pro.
  • GPT-5 — Information on OpenAI’s GPT-5 series, including the Pro and Codex variants.
  • Digital Services Act — EU regulation mentioned in the context of ARCOM’s strategy.
  • Nvidia — Background on Nvidia’s financial performance and role in AI.
  • TikTok digital well-being — General information on TikTok’s features, including well-being tools.

106 words

Radar Profile

The radar profile shows a balanced distribution across information quantity, quality, technical level, and reliability, with a slight emphasis on quantity and reliability. This suggests a video that is informative and generally reliable, but not highly technical or deeply analytical.

Reliability 6/10

💬 Sur les 0 commentaires analysés, aucune tendance n'est disponible.