Nvidia dominates GPUs, Google bets on TPUs, and Microsoft races into custom chips—while hallucination‑prone LLMs fuel a risky AI spending boom and world models quietly prepare to move AI into the physical world.
🎯 DISCOVER
- How Google’s TPU‑powered Gemini 3 challenges Nvidia’s GPU dominance by shifting toward autonomous task completion.
- Why leading AI models still hallucinate at extreme rates—and why pushing for more accuracy can actually worsen errors.
- How trillions in AI infrastructure commitments could fail to deliver ROI and why investors are increasingly uneasy.
- How ASIC chip specialization may squeeze Nvidia as Microsoft, Google, and Amazon double down on custom architectures.
- Why hybrid AI and world models could reset robotics, manufacturing, and real‑world planning.
- How Microsoft’s fast‑follow system strategy and access to Broadcom chip IP give it leverage even over frontier labs.
🔍 TIMESTAMPS
00:00 Intro — Wins, Fails and the Future of AI
00:00:38 Nvidia’s Record Quarter and GPU Lead
00:01:08 Google’s Gemini 3 and TPU Advantage
00:02:03 Hallucinations: Accuracy vs Confidently Wrong
00:03:51 The AI Investment Bubble and ROI Concerns
00:04:42 Trillion-Dollar Infrastructure Commitments
00:07:58 Future Shift: World Models and Hybrid AI
00:10:55 Robotics, Fast-Follow Strategies and Chip Competition
00:14:48 Key Trends to Watch for 2026 and What No One Knows Yet
👍 Found value? Like, comment and share so more leaders see this video!
📚 SOURCES & DATA
- AI Reality Check $400B Market: https://youtu.be/32-uTzeRHTw
- Nvidia Q3 2025 earnings and AI chip sales: https://www.wsj.com/tech/ai/nvidia-earnings-q3-2025-nvda-stock-9c6a40fe
- Google Gemini 3 AI launch and TPU innovation: https://www.reuters.com/business/media-telecom/google-launches-gemini-3-embeds-ai-model-into-search-immediately-2025-11-18/
- Top AI chips comparison: Nvidia, Google TPUs, AWS Tranium: https://www.cnbc.com/2025/11/21/nvidia-gpus-google-tpus-aws-trainium-comparing-the-top-ai-chips.html
- DeepSeek worst hallucination rates in AI: https://www.nature.com/articles/s43856-025-01021-3
- Gemini 3 reliability and hallucination benchmark: https://the-decoder.com/gemini-3-pro-tops-new-ai-reliability-benchmark-but-hallucination-rates-remain-high/
- Tech must deploy $500B by 2030: https://www.bain.com/insights/topics/technology-report/
- Meta spending $600B on AI data centers: https://www.reuters.com/business/meta-plans-600-billion-us-spend-ai-data-centers-expand-2025-11-07/
- China’s $70B AI data center investment: https://www.goldmansachs.com/insights/articles/chinas-ai-providers-expected-to-invest-70-billion-dollars-in-data-centers-amid-overseas-expansion
- Additional DeepSeek and AI data center investment context: https://www.nytimes.com/2025/02/08/technology/deepseek-data-centers-ai.html
- DeepSeek impact on booming data center market: https://www.cnbc.com/2025/02/14/how-chinas-deepseek-could-boost-the-already-booming-data-center-market.html
- AI Chips & China:https://youtu.be/6Juw5ZiL3Es
- US-China Truce: Key Issues Unresolved: https://youtu.be/3F_zislTiKI
- Microsoft, Nvidia, Anthropic partnership details: https://www.morningstar.com/news/marketwatch/2025111881/microsoft-nvidia-and-anthropic-ink-45-billion-deal-that-furthers-a-big-critique-of-the-ai-trade
- World Models AI research comeback: https://www.quantamagazine.org/world-models-an-old-idea-in-ai-mount-a-comeback-20250902/
- MIT on LLMs understanding real world: https://news.mit.edu/2025/can-large-language-models-figure-out-real-world-0825
Full list: https://youtu.be/nIp8HGLQeNI
","author_name":"Neeta Bidwai"}