The AI War: OpenAI Ads & Sora 2, Grok Partners With US Government & Google’s Ad Business is at Risk w/ Dave Blundin, Salim Ismail, & Alex Wissner-Gross | EP #198
Timestamps are as accurate as they can be but may be slightly off. We encourage you to listen to the full context.
Podcast Summary
This Moonshots podcast episode features Peter Diamandis, Dave Blunden, Salim Ismail, and Alex discussing the rapid convergence of AI technologies in early 2025. The hosts explore how we're transitioning from algorithmic content selection to algorithmic content generation, examining breakthrough video and audio generation tools like Sora 2 and Meta's Vibes app. (09:01) They delve into the fierce competition between AI labs, with Anthropic's Claude Sonnet 4.5 achieving new coding milestones while OpenAI expands into advertising and commerce integration. The discussion spans critical infrastructure topics including massive data center investments, energy demands, and the geopolitical implications of AI supremacy.
- Main Theme: The podcast examines how AI is rapidly transforming from experimental technology to practical tools that are reshaping video creation, software development, energy infrastructure, and global economic systems in real-time.
Speakers
Peter Diamandis
Founder and Executive Chairman of XPRIZE Foundation and co-founder of Singularity University. Diamandis recently ran his Abundance Longevity Summit featuring 50 of the world's top longevity scientists and entrepreneurs. He's a bestselling author and leading voice in exponential technologies, focused on extending healthy human lifespan and solving grand challenges.
Dave Blunden
Co-founder of Link Ventures, operating out of Harvard and MIT ecosystems. Blunden specializes in early-stage AI companies and has a track record of identifying breakthrough technologies. He was instrumental in taking D-Wave public through a SPAC and focuses on building AI-native startups with founding teams of "bona fide best friends."
Salim Ismail
Former head of Singularity University and expert in exponential organizations. Ismail is known for his ability to spot exponential trends before they become mainstream and helps organizations transform to handle exponential growth. He runs transformation workshops and speaks globally on organizational adaptation to rapid technological change.
Alex
AI researcher and expert focused on frontier model capabilities and AGI timelines. Alex has published research on AI alignment and power-seeking behaviors in AI systems. He closely tracks the development of frontier models and provides analysis on the path to artificial general intelligence and superintelligence.
Key Takeaways
Master the New Creative Tools Before Your Competition
The transition from content selection algorithms to content generation algorithms is happening now, with tools like Sora 2 and Meta's Vibes app making professional-quality video creation accessible to anyone. (10:11) These tools aren't just impressive demos—they're becoming practical business tools that can create viral content in minutes. The most successful professionals will be those who integrate these AI-powered creative capabilities into their workflows immediately, rather than waiting for the technology to mature further.
Prepare for the Autonomy Revolution in Work
Claude Sonnet 4.5 can now work autonomously for 30+ hours straight on complex coding tasks, marking a dramatic leap from previous models that could only work independently for 7 hours or less. (20:48) This represents a hyper-exponential growth curve in AI autonomy that suggests most knowledge work will be superhuman-level within 6-12 months. Professionals need to start thinking now about how to collaborate with AI systems that can work independently for days or weeks, fundamentally changing project management and workflow design.
Invest in Real-World Application Over Theory
The new APEX and GDPVal benchmarks show AI models completing real-world professional tasks up to 100 times faster and cheaper than human experts. (44:58) Rather than focusing on theoretical AI capabilities, successful professionals should identify specific, measurable tasks in their industry and test how AI can enhance or replace these functions. The companies winning today are those creating practical benchmarks for their specific use cases and iterating rapidly on real business problems.
Embrace the Compute-First Mindset
Sam Altman's call for 10 gigawatts of compute and Meta's $600 billion AI infrastructure commitment signal that abundant compute access will determine competitive advantage. (71:54) Business leaders who aren't securing dedicated compute resources now will find themselves unable to compete when real-time AI application generation and autonomous agents become standard. The scarcity isn't just in chips—it's in guaranteed access to the computational power needed for AI-native business operations.
Prepare for Exponential Blindness in Your Industry
Expert predictions consistently underestimate exponential growth by orders of magnitude, as demonstrated by solar energy forecasts that predicted linear growth while reality showed hockey-stick exponentials. (99:25) The same cognitive blind spots that caused energy experts to miss solar's exponential curve are affecting AI predictions today. Professionals must actively fight their intuition about linear progress and instead assume that current AI capabilities will improve 10-100x faster than expert consensus predictions suggest.
Statistics & Facts
- Global solar capacity increased from 40 gigawatts in 2010 to almost 3 terawatts in 2025, making it the fastest growing energy source in history. (99:27) This represents a 75x increase in just 15 years, demonstrating how expert predictions consistently underestimate exponential technology adoption curves.
- 32 gigabytes of RAM that costs $100 today would have cost approximately $150 million in the mid-1980s, while a 4-terabyte hard drive would have cost $1.26 billion. (89:38) This demonstrates the extraordinary exponential deflation in computing costs that continues to accelerate AI capabilities.
- If every US vehicle performed as well as Waymo, it would prevent 33,000-39,000 deaths annually, representing a 90% reduction in fatal crashes. (114:22) Additionally, 50% of all court cases in the US currently involve car accidents, highlighting the massive societal impact autonomous vehicles will have beyond just safety.
