Ces jeux IMPOSSIBLES créés par des IA en quelques clics (on essaie !)

Ces jeux IMPOSSIBLES créés par des IA en quelques clics (on essaie !)

These IMPOSSIBLE games created by AIs in a few clicks (we try!)

🎙 Frandroid 👥 625K 📅 October 20, 2026 ⏱ 11 min 👁 610 📄 news review 🧭 2026-10-20
Available in: English (current) Français

Keywords

pass-throughrecompilationClaudeRustmodding

Summary

The video explains recent viral videos where characters from one game appear in another (e.g., Mario in Elden Ring, Minecraft in Skyrim) are not fakes but are actually playable mods. It details three technical approaches: the pass-through method where two games run simultaneously and communicate via plugins, the use of decompiled game engines like that of Mario 64, and rewriting a game’s engine in Rust with AI assistance, as done for Call of Duty and Skate 3. The presenter demonstrates the SkyCraft mod personally, highlighting its integration and functionality. The role of AI, particularly code-generating models like Claude, is emphasized as a catalyst, enabling complex mods to be created in hours. The video also discusses the potential for porting console exclusives to PC, citing examples like Infamous and Fable 2, and addresses legal and copyright concerns, noting corporate reactions. It concludes that while these are prototypes with limitations, they signal a new era for modding and game preservation.

158 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides substantial value by demystifying a trending internet phenomenon and grounding it in concrete technical explanations. The argumentation is strong: the presenter tests a real mod (SkyCraft) and shows its inner workings, referencing the GitHub repository and the method of pass-through. He also explains the role of community reverse-engineering and how AI accelerates the process, citing specific projects and timelines. The explanations are logical and accessible, and the video acknowledges both possibilities and limitations, which enhances its credibility.

Scientific Rigor, Source Quality, Title Accuracy

The video demonstrates scientific rigor by citing real projects, providing links to a GitHub repository and articles from the same outlet, and by distinguishing between verified demonstrations and speculative claims. It explains technical concepts accurately, such as pass-through and recompilation, without oversimplification. The title is well-matched to the content, as it promises a practical exploration of AI-created games, which the video delivers. No significant discrepancies between title and content were observed.

166 words

Title / Content Match

The title accurately reflects the content: it promises impossible games created by AI and the video indeed demonstrates several such mods, backed with explanations and examples.

Quality & Reliability

8/10

The video is well-documented, referencing specific GitHub repositories and real projects, and is produced by a reputable tech media outlet. It distinguishes clearly between speculation and verified demonstrations, and acknowledges legal and technical limitations.

Chapters

Cited Sources

External References

Contribution & Novelties

The video provides a clear and structured explanation of how AI-assisted game modding works, illustrating with real examples and a hands-on test. Its novelty lies in synthesizing the technical methods (pass-through, engine extraction, full rewrites) and contextualizing them within the current AI boom, while also addressing legal implications. It offers a practical perspective for viewers unfamiliar with the underlying code.

Pour aller plus loin :

124 words

Radar Profile

The radar profile shows high scores in information quantity, quality, and reliability, with a moderate technical level. This indicates a well-researched, accessible video that provides substantial useful information without making the content too specialized. The overall high scores suggest the video is both engaging and trustworthy.

Reliability 8/10