Skip to main content
All-In with Chamath, Jason, Sacks & Friedberg

Can the AI Industry Regulate Itself? Stripe Wants PayPal, China Catches Up, NY Bans Datacenters

89 min episode · 3 min read

Episode

89 min

Read time

3 min

Topics

Productivity, Startups, Leadership

AI-Generated Summary

Key Takeaways

  • AI Self-Regulation Framework: David Sacks outlines five conditions for a viable AI SRO: broad representation including startups and open source, review limited to true frontier models only, focus exclusively on catastrophic risks like cyber and CBRN threats, voluntary operation before becoming mandatory, and functioning as a substitute for new government agencies rather than an additive layer. Without these guardrails, the SRO becomes an opening bid for deeper regulatory capture.
  • Regulatory Capture Playbook: Anthropic pursues a state-by-state strategy of incrementally tightening AI regulations rather than aligning around a single national framework, per a Politico investigation. Each new state adoption introduces stricter rules than the previous one, creating a ratchet effect. Anthropic also funds NGOs opposing data center construction, despite compute scarcity being the primary constraint on its own revenue growth.
  • Payments Consolidation Thesis: Stripe, Block, and private equity firm Advent are jointly bidding roughly $53-60B for PayPal, which peaked at $322B and now trades near $40B. The strategic logic: combining Stripe's 2T annual merchant volume, PayPal's 439M consumer accounts, Block's point-of-sale infrastructure, Cash App, Venmo, Braintree, and stablecoin assets creates a viable end-to-end competitor to Visa and Mastercard on proprietary rails.
  • AI Token Cost Arbitrage: Input token pricing varies by 100x across providers: Claude charges approximately $56 per million tokens, GPT-4 around $26, Grok roughly $1.50, and leading Chinese models near $0.50. Ramp data shows enterprise token spend grew 21x in one year, not 21%. CFOs lack visibility into this spend, creating earnings risk. Switching to lower-cost models for non-frontier tasks captures most of the savings with minimal quality loss.
  • Data Center Energy Crisis: The US faces an energy deficit equivalent to 2.5 times California's total consumption by 2050. PJM, the utility serving 13 states, needed 7-8 gigawatts in a recent capacity auction and received roughly 156 megawatts. Approximately 40% of planned data center projects are being cancelled or paused. New York's statewide hyperscale moratorium effectively creates a 5-year delay when accounting for permitting and construction ramp-up timelines.

What It Covers

The All-In hosts debate Demis Hassabis's proposal for an AI self-regulatory organization modeled after FINRA, analyze Stripe's $53B bid for PayPal alongside Block and Advent, examine New York's data center moratorium, cover Apple's trade secret lawsuit against OpenAI, and discuss a breakthrough enzyme that reverses skin aging by 40 years.

Key Questions Answered

  • AI Self-Regulation Framework: David Sacks outlines five conditions for a viable AI SRO: broad representation including startups and open source, review limited to true frontier models only, focus exclusively on catastrophic risks like cyber and CBRN threats, voluntary operation before becoming mandatory, and functioning as a substitute for new government agencies rather than an additive layer. Without these guardrails, the SRO becomes an opening bid for deeper regulatory capture.
  • Regulatory Capture Playbook: Anthropic pursues a state-by-state strategy of incrementally tightening AI regulations rather than aligning around a single national framework, per a Politico investigation. Each new state adoption introduces stricter rules than the previous one, creating a ratchet effect. Anthropic also funds NGOs opposing data center construction, despite compute scarcity being the primary constraint on its own revenue growth.
  • Payments Consolidation Thesis: Stripe, Block, and private equity firm Advent are jointly bidding roughly $53-60B for PayPal, which peaked at $322B and now trades near $40B. The strategic logic: combining Stripe's 2T annual merchant volume, PayPal's 439M consumer accounts, Block's point-of-sale infrastructure, Cash App, Venmo, Braintree, and stablecoin assets creates a viable end-to-end competitor to Visa and Mastercard on proprietary rails.
  • AI Token Cost Arbitrage: Input token pricing varies by 100x across providers: Claude charges approximately $56 per million tokens, GPT-4 around $26, Grok roughly $1.50, and leading Chinese models near $0.50. Ramp data shows enterprise token spend grew 21x in one year, not 21%. CFOs lack visibility into this spend, creating earnings risk. Switching to lower-cost models for non-frontier tasks captures most of the savings with minimal quality loss.
  • Data Center Energy Crisis: The US faces an energy deficit equivalent to 2.5 times California's total consumption by 2050. PJM, the utility serving 13 states, needed 7-8 gigawatts in a recent capacity auction and received roughly 156 megawatts. Approximately 40% of planned data center projects are being cancelled or paused. New York's statewide hyperscale moratorium effectively creates a 5-year delay when accounting for permitting and construction ramp-up timelines.
  • Extracellular Matrix Age Reversal: Calico and Revel Pharma used AlphaFold combined with five rounds of directed protein evolution to engineer a novel enzyme that degrades CML, the primary advanced glycation end product accumulating between cells with age. Applied to donated skin from patients over 70, the enzyme eliminated 55% of CML, reverting tissue markers to those of a 31-year-old. Cosmetic topical application is the likely first commercial pathway.

Notable Moment

Sacks reveals he personally advised Demis Hassabis on the SRO proposal, outlining five specific conditions for it to avoid becoming a regulatory capture vehicle. He warns that without explicit preemption language written into law, the SRO will simply serve as Anthropic's opening bid toward an FAA-style permission-based regime requiring years-long model approval timelines.

Know someone who'd find this useful?

Episode Transcript

Everybody. Welcome back to the world's greatest podcast, the number one podcast, your favorite podcast, the all in podcast. I'm Jason Calacanis, the world's greatest moderator. With me, of course, Shamath Palihapatidis. Great. Great. Great. Great great great great great grandchild, Jason was, of a hooker and You saw that from France? And a purse snatcher. This this comes from, like, a history of, France. Some history said they let you add up this was in this was the deal in 1719, Sacks. If you're a prisoner in a in Paris, you were offered your freedom on the condition that you marry a prostitute and move to the great state of Louisiana. What are you saying? It's actually taking the deal? It explains, a certain of your proclivities, Jayco. I I thought you were asking me to see if they would extend the rule for you. No. I'm saying that your great great great great great great grandmother was a hooker. I'm great. That's what I'm saying very explicitly. She was, like We never we never spent time in a prison. Also, of course, David Freiburg is here. How are you doing, brother? Living the dream. Yeah. Good to be back. Missed you guys last week. How was Brad? How did he fill in? That was great. Yeah. Yeah. He was great. Count victory lap? Yeah. The little victory lap. We played Chariots of Fire. And, how was your special time at blank and your time next week at blank? Thanks for thanks for having me on your show, Diego. Alright. We got a full docket today. Lots of stories. Let's start with, DeepMind's Demis Hassabis just dropped an AI regulation proposal, and it's, it's pretty popular with the boys. In an x article, Demas called for a US led international AI standards body. Proposal is modeled after FINRA, the Financial Industry Regulatory Authority. That's a self regulatory body, and this would be federally overseen but industry funded and run by independent technological experts. Frontier Labs would submit their models thirty days before release, and it would be voluntary initially, then mandatory at some point. The models would be assessed on risk to cybersecurity, national security, biological threats, and other high risk domains. Benchmarks would be updated quarterly, and the body can coordinate a slowdown in development if the situation demands it. I guess that would be if, there was a cyber risk, etcetera. Looks like, on the positive side, we have Elon who said it was thoughtful, Sam at OpenAI, Jack Clark at Anthropic, Sundar, Satya, Jack Dorsey from Block, the Carlson brothers. So, Freeberg, Friberg, your thoughts on Oh, well, way to read it way to really, put in the effort today. Go ahead, Jason. Could could you just pass it to me, James, to take care of it? Let me let me What do you want me to do? Oh, my god. What an incredible topic. Give me the Alright. Here. Let me do it. You want …

Get the full transcript (16,720 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all All-In with Chamath, Jason, Sacks & Friedberg transcripts →

You just read a 3-minute summary of a 86-minute episode.

Get All-In with Chamath, Jason, Sacks & Friedberg summarized like this every Monday — plus up to 2 more podcasts, free.

Pick Your Podcasts — Free

Keep Reading

Books, tools, and gear mentioned in this episode

SignalCast may earn commission on purchases via these links. As an Amazon Associate, SignalCast earns from qualifying purchases.

Tools

  • by Anthropic

    Claude charges approximately $56 per million tokens, GPT-4 around $26, Grok roughly $1.50
  • by OpenAI

    Claude charges approximately $56 per million tokens, GPT-4 around $26, Grok roughly $1.50
  • Claude charges approximately $56 per million tokens, GPT-4 around $26, Grok roughly $1.50
  • by DeepMind

    Calico and Revel Pharma used AlphaFold combined with five rounds of directed protein evolution to engineer a novel enzyme

Products

  • combining Stripe's 2T annual merchant volume, PayPal's 439M consumer accounts, Block's point-of-sale infrastructure, Cash App, Venmo, Braintree, and stablecoin assets
  • combining Stripe's 2T annual merchant volume, PayPal's 439M consumer accounts, Block's point-of-sale infrastructure, Cash App, Venmo, Braintree, and stablecoin assets
  • combining Stripe's 2T annual merchant volume, PayPal's 439M consumer accounts, Block's point-of-sale infrastructure, Cash App, Venmo, Braintree, and stablecoin assets

company

  • Stripe's $53B bid for PayPal alongside Block and Advent
  • Stripe's $53B bid for PayPal alongside Block and Advent, examine...PayPal, which peaked at $322B and now trades near $40B
  • Stripe's $53B bid for PayPal alongside Block and Advent
  • Stripe's $53B bid for PayPal alongside Block and Advent
  • Calico and Revel Pharma used AlphaFold combined with five rounds of directed protein evolution to engineer a novel enzyme
  • Calico and Revel Pharma used AlphaFold combined with five rounds of directed protein evolution to engineer a novel enzyme

More from All-In with Chamath, Jason, Sacks & Friedberg

We summarize every new episode. Want them in your inbox?

Similar Episodes

Related episodes from other podcasts

Explore Related Topics

This podcast is featured in Best Tech Podcasts (2026) — ranked and reviewed with AI summaries.

Read this week's Startups & Product Podcast Insights — cross-podcast analysis updated weekly.

You're clearly into All-In with Chamath, Jason, Sacks & Friedberg.

Every Monday, we deliver AI summaries of the latest episodes from All-In with Chamath, Jason, Sacks & Friedberg and 192+ other podcasts. Free for one show.

Start My Monday Digest

No credit card · Unsubscribe anytime