Uploaded March 2026 | Updated September 2026, 1 week ago
The AI Scientist Who Called Nvidia Early Is Back — And The Questions Are Different This Time
When Dr. Craig Kaplan first joined me, Nvidia was still being debated. He called it from first principles when Wall Street was still asking whether the run-up made sense.
He returns this Wednesday for his fourth appearance — and the landscape has changed completely.
Gemini 3.1 Pro. Claude Opus 4.6 and Sonnet 4.6. GPT-5.4. Grok 4.20. DeepSeek V4 imminent. Claude Code. Seedance. A one-million token context window becoming the new standard. Agents finally moving from demo to deployment.
Craig is Founder and CEO of SuperIntelligence.com and iQ Company — an AI scientist since the 1980s and a retired hedge fund manager. He evaluates these systems with a rigor most investors and commentators simply do not have.
We are going deep on:
— Which models are actually ahead right now and why benchmarks may be hiding the real story
— Context windows: why the jump to one million tokens is not a spec upgrade but a fundamental shift in what AI can do for a business
— Claude vs Gemini vs GPT vs Grok — a real compare and contrast, not marketing
— DeepSeek V4 and why China's open-weight strategy may be the most underappreciated geopolitical AI move happening right now
— Claude Code, Seedance, and the application layer that is finally arriving
— Grok 4.20's four-agent parallel architecture and why it is technically novel
— The transformer architecture question nobody wants to ask: are we scaling toward superintelligence on a foundation we do not mechanistically understand?
— Recursive self-improvement — the threshold every safety researcher agrees is the critical one — and how close any system actually is to it
— The Mag 7 updated scorecard with real conviction changes
— Robotics, Nvidia, the infrastructure layer, and where the next ten-year money actually is
— How family offices should build an AI investment framework from scratch in March 2026
This is not a hype session. Craig brings scientific depth and investment discipline to a conversation most people are having with neither.
Wednesday, March 11th | 4:00 PM ET | Live
Register below. Free to attend.
---
Family offices managing generational wealth need more than advisors — they need a network operating 72 hours ahead of the curve. SFO Continuity delivers sovereign AI blueprints, peer roundtables, and proprietary frameworks forged inside 100+ billionaire ecosystems. Membership is limited to 100 family offices.
Learn more: angelorobles.com/membership
#FamilyOffice #AI #ArtificialIntelligence #Nvidia #Claude #Gemini #GPT5 #DeepSeek #AIAgents #Investing #SFOContinuity #Superintelligence #FamilyOfficeTV #AngeloRobles #AIInvesting #angelorobles
The AI Scientist Who Called Nvidia Early Is Back — And The Questions Are Different This Time
When Dr. Craig Kaplan first joined me, Nvidia was still being debated. He called it from first principles when Wall Street was still asking whether the run-up made sense.
He returns this Wednesday for his fourth appearance — and the landscape has changed completely.
Gemini 3.1 Pro. Claude Opus 4.6 and Sonnet 4.6. GPT-5.4. Grok 4.20. DeepSeek V4 imminent. Claude Code. Seedance. A one-million token context window becoming the new standard. Agents finally moving from demo to deployment.
Craig is Founder and CEO of SuperIntelligence.com and iQ Company — an AI scientist since the 1980s and a retired hedge fund manager. He evaluates these systems with a rigor most investors and commentators simply do not have.
We are going deep on:
— Which models are actually ahead right now and why benchmarks may be hiding the real story
— Context windows: why the jump to one million tokens is not a spec upgrade but a fundamental shift in what AI can do for a business
— Claude vs Gemini vs GPT vs Grok — a real compare and contrast, not marketing
— DeepSeek V4 and why China's open-weight strategy may be the most underappreciated geopolitical AI move happening right now
— Claude Code, Seedance, and the application layer that is finally arriving
— Grok 4.20's four-agent parallel architecture and why it is technically novel
— The transformer architecture question nobody wants to ask: are we scaling toward superintelligence on a foundation we do not mechanistically understand?
— Recursive self-improvement — the threshold every safety researcher agrees is the critical one — and how close any system actually is to it
— The Mag 7 updated scorecard with real conviction changes
— Robotics, Nvidia, the infrastructure layer, and where the next ten-year money actually is
— How family offices should build an AI investment framework from scratch in March 2026
This is not a hype session. Craig brings scientific depth and investment discipline to a conversation most people are having with neither.
Wednesday, March 11th | 4:00 PM ET | Live
Register below. Free to attend.
---
Family offices managing generational wealth need more than advisors — they need a network operating 72 hours ahead of the curve. SFO Continuity delivers sovereign AI blueprints, peer roundtables, and proprietary frameworks forged inside 100+ billionaire ecosystems. Membership is limited to 100 family offices.
Learn more: angelorobles.com/membership
#FamilyOffice #AI #ArtificialIntelligence #Nvidia #Claude #Gemini #GPT5 #DeepSeek #AIAgents #Investing #SFOContinuity #Superintelligence #FamilyOfficeTV #AngeloRobles #AIInvesting #angelorobles
