Podcast Summary

Trends with Friends Ep: "Google vs OpenAI: Code Red, Gemini, and the AI Arms Race"

Date: December 9, 2025
Host: Howard Lindzon
Guests: Michael (ex-Goldman, AI newsletter author), Roy Rubin (founder of Magento, R Squared Ventures)

Episode Overview

This episode dives into the escalating AI arms race between Google and OpenAI, dissecting the significance of Google’s Gemini 3 launch and the resulting “Code Red” at OpenAI. The discussion covers why Google may now be the frontrunner, what this means for users, and how companies like Apple, Amazon, and Nvidia are jockeying for position. The hosts also explore deeper questions about AI product stickiness, business models, and implications for e-commerce platforms.

Key Discussion Points & Insights

1. From OpenAI Hype to Google’s “Code Red” Moment

• Backdrop: While OpenAI captured the spotlight and user growth with ChatGPT, Google played the long game, continuing to invest and innovate—culminating in the Gemini 3 model release.
• Gemini 3 Release: This leap, together with significant hardware and cloud announcements, has caused a shift, leading OpenAI’s Sam Altman to reportedly declare an internal “Code Red.”
  • “[Gemini 3] just totally beat everyone's expectations relative to the best from OpenAI which was GPT 5.1.” — Michael (10:10)
• Competition Mirrors Tech History:
  • “It’s a movie we’ve seen before. We saw Microsoft versus Netscape in the mid-90s…” — Michael (07:05)
• OpenAI’s Challenges: OpenAI faces research slowdowns, talent poaching (notably by Meta), competition in AI chips, and the pressures of running their own cloud infrastructure after separation from Microsoft.

2. Network Effects, User Behavior, and Product “Stickiness”

• Google’s Strategic Advantage:
  • Google’s 25-year head start in data collection (via Search, Maps, Gmail, YouTube) means more data for training, massive distribution, and deeply ingrained user habits.
  • “You don't have to buy a new device… it's just in the same browser. That’s what I did with my mom. She didn't even know she was using Gemini.” — Michael (31:08)
• The Declining Dominance of ChatGPT:
  • Several anecdotal signs and data showing users migrating from ChatGPT to Gemini for daily tasks, especially since Gemini’s AI heavily integrates into the standard Google search experience.
  • “It’s been days since I’ve used the ChatGPT app on my Mac. This used to be an hourly occurrence.” — Roy (15:51)
• UI and Distribution Are Critical:
  • “UI matters.” — Michael (12:30), emphasizing that even the best AI model loses if the interface isn’t familiar or easy.

3. The Economics and Business Models Behind AI

• Revenue Models:
  • OpenAI, while “the most successful subscription driven business on the planet,” still must compete with Google and Apple, who have powerful ad, ecosystem, and device moats.
  • “They don’t have a business model outside of subscriptions…that’s going to decline.” — Roy (21:07)
  • OpenAI’s pause on ad business to focus on narrowing Gemini’s lead is seen as high-risk.
• Apps and Platforms:
  • There is tension between building first-party solutions (apps) and facilitating a thriving third-party ecosystem. OpenAI is “starting incredibly early” in trying to find killer app revenue, unlike Microsoft, Apple, or Google historically.
  • “If you’re a platform player you want to leave holes for others to plug into… OpenAI is not leaving a lot of room for others.” — Roy (21:27)

4. The Power Players: Apple, Nvidia, Anthropic, and Amazon

• Apple’s Position:
  • Stands apart, quietly leveraging ecosystem and device control to benefit regardless of shifts in AI model leadership.
  • “Maybe not the most innovative company anymore, but it’s stronger than ever.” — Howard (27:22)
  • Likely to use Gemini for Siri—a big boost for Google’s distribution.
• Nvidia’s Role:
  • Jensen Huang (CEO) praised Google’s progress—a high-stakes, delicate act as Nvidia must remain essential to all sides in the arms race.
• Amazon & E-commerce:
  • Amazon’s AWS is the gateway for mainstream business AI, now investing heavily in agentic tech, their own chips, and drawing boundaries against AI platforms scraping their e-commerce data.
  • “Amazon is basically, it’s really a shot across the bow for OpenAI, for Google, for anybody else who wants to have their AI becoming the key way to shop for people.” — Michael (45:42)
• Anthropic:
  • Building a B2B API-first business, with investment from all the tech giants and prepping for a potentially massive IPO in 2026.

5. E-commerce & the AI Interface Dilemma

• Current Limitations:
  • Shopping through ChatGPT, Perplexity, or similar interfaces is “transactional, one-to-one, and leaves a lot to be desired” vs. dedicated merchant sites.
  • “There has to be parity at some point… but right now, it’s all noise.” — Roy (50:43, 54:12)
• Walled Gardens:
  • Fragmentation means users may be trapped in Gemini, OpenAI, Perplexity, or cloud “worlds”—all risking a future with incompatible experiences and “different types of truths.”
  • “Are those worlds going to talk to each other? Because I could see a world where that’s all gated.” — Roy (53:41)
• Shopify, Magento, etc.:
  • No existential threat yet; the core e-commerce experience is still superior outside AI interfaces, but things could change with faster agent tech or if AI platforms open up for deeper integrations.

6. The New Attention Wars: Video, Prediction Markets, and AI

• YouTube’s Dominance:
  • “If we get down to it, video… YouTube… you can’t win that.” — Howard (35:14)
• Prediction Markets as News:
  • Howard sees platforms like Kalshi and Polymarket as the next evolution in “news”, disrupting the status quo of argument and opinion on social networks by betting on outcomes instead.

Notable Quotes & Memorable Moments

• On Google’s Quiet Strategy:
  • “They’ve been working on it for three years, actually years longer. Google, you know, actually came up with the transformer paper in 2017 that gave us the T in GPT.” — Michael (08:52)
• On User Habit Inertia:
  • “Most people… have not downloaded a new browser in a decade or more.” — Michael (32:44)
• On OpenAI’s Weakness:
  • “You don’t have Chrome, you don’t have Gmail… They don’t have a business model outside of subscriptions.” — Roy (21:04)
• On AI at ‘the Edge’:
  • “If we can get AI at the edge, it’s all we need.” — Roy (30:26)
• On FSD as New Luxury:
  • “The new luxury is time. If you can get my time back, I’ll pay anything for it.” — Roy (39:42)

Timestamps for Important Segments

• [06:51] “Code Red” at OpenAI, Google’s AI resurgence
• [10:36] Gemini 3: Google’s vertical stack, TPUs, and competition with Nvidia
• [12:30] Network effects, data, & UI in AI adoption
• [15:51] User behavior shifting: Gemini vs ChatGPT
• [19:12] OpenAI revenue challenges, business models, and ad pauses
• [26:31] Apple’s role, partnerships, and strategy
• [32:44] AI browsers, friction, and user inertia
• [35:14] The enduring power of YouTube and Google’s stack
• [38:01] Tesla/FSD as “new luxury”; Google’s Waymo and autonomy
• [42:04] Amazon’s strategic moves in AI and E-commerce
• [50:00] E-commerce perspective: Shopify, Magento, agentic commerce limitations
• [53:41] The coming walled garden problem in AI “worlds”

Takeaways for Listeners

• The “AI arms race” is now a heavyweight battle between Google, OpenAI, and a handful of megacaps—distribution, ecosystem, and data depth are just as critical as AI model quality.
• User habits and distribution channels (search, browsers, app stores) trump short-term technological leaps in the AI world.
• OpenAI faces existential business pressures beyond just tech—lacking deep platform integration and roadmap clarity compared to Google, Apple, and Amazon.
• E-commerce via general AI interfaces falls woefully short—for now—suggesting incremental rather than revolutionary disruption.
• The next two to three years will be highly experimental in AI, with rapid change and opportunity, but also massive risk for also-rans and business models not built on solid ground.

Ending Note

“The most important thing though is how do regular people—Fred’s mom, etc.—what are their behaviors? And that’s where the real opportunity is.” — Michael (55:45)