GPT-5.1 vs GEMINI 3 PRO (and more): Is the AI Bubble about to POP? (LLM Council Pus DEBATE)

🤯 IS THE AI BUBBLE REAL? 🫧
The question every tech expert, investor, and builder is asking.

Instead of listening to the endless human debates, I put the world's most advanced Large Language Models - including the new **GPT-5.1**, **Gemini 3 Pro**, **Claude Opus 4.5**, and **DeepSeek v3.2 Speciale** -in a council to settle the score.

The results were fascinating. The models used real-time web search results and then went through a brutal peer-review stage. As you'll see in the video, **the big SOTA models were a bit harsh on it!**

Watch as they analyze the market, rank each other's responses, and ultimately deliver a single, synthesized final verdict on the severity and timing of a correction.

**🛠️ The Tech Behind the Council**
This entire system is built on "LLM Council Plus," a fork of the new, highly-discussed open-source project by **Andrej Karpathy** (the llm-council repository).

**CHAPTERS / JUMP TO:**
0:00 - Welcome to the LLM Council
0:15 - The Question: Is the AI Bubble Real?
0:30 - Web Search & Source Articles
0:46 - Stage 1: Individual LLM Deliberation
1:07 - Stage 2: Peer Rankings & Brutal Reviews
1:35 - Visual Leaderboard & Top Performer
1:53 - The CHAIRMAN's Final Verdict (The Answer!)

**GET INVOLVED:**
➡️ Like this video and subscribe for more LLM deep dives!
➡️ Leave a comment: Which model do you think won the debate?
➡️ Check out the original Karpathy project: https://github.com/karpathy/llm-council
➡️ My LLM Council Plus Fork/Code: COMING SOON 🚀

Видео GPT-5.1 vs GEMINI 3 PRO (and more): Is the AI Bubble about to POP? (LLM Council Pus DEBATE) канала Gen AI Spotlight