Podcast Summary: Wisdom of the $TAO — The Future is Decentralized AI
This Week in Startups with Jason Calacanis
Date: March 6, 2026
Overview
In this episode, Jason Calacanis explores the cutting edge of decentralized artificial intelligence, focusing on Bittensor ($TAO) — a project that melds blockchain with open AI infrastructure. Partnering with leading investors and founders, Jason investigates how programmable crypto mining, tokenized AI subnets, and community-driven development are shaping the future of AI and decentralization. The discussion covers investments, emerging products, open agent platforms, and the broader implications for startups, founders, and society at large.
Key Topics & Discussion Points
1. The Intersection of Crypto and AI (03:56 – 06:49)
- Jason’s Pivot Advice: Reflects on advice given for crypto founders to pivot to AI and acknowledges the rise of regulatory clarity in the US.
- “I am not saying pivot from AI to crypto. That would be foolish, but I would say keep an eye on crypto…” (04:36)
- Regulatory Progress: Recent interviews with the heads of the SEC and CFTC point to a more legitimate US crypto environment.
2. Introduction to Bittensor and the $TAO Network (05:58 – 09:52)
- Guest: Mark Jeffrey (Stillcore Capital)
- What is Bittensor?: A blockchain that incentivizes the development of AI products via programmable mining—subsidizing innovation akin to Bitcoin mining, but for useful AI.
- “It’s basically taking bitcoin’s mining side and making that programmable on a new chain.” (05:58)
- Tokenomics: $100M/year issued in TAO subsidies to AI subnet innovators.
- Comparison to Early Bitcoin: Mining rewards based on utility and talent rather than pure computational “busy work.”
- “Bittensor does for stranded talent what Bitcoin did for stranded energy.” (09:52)
3. Examining Decentralized AI Innovations (08:03 – 12:56)
- Ridges (Subnet 62): Open-source Vibe coding copilot competitive with Claude/Cursor, built for $10M in emissions vs. Cursor's $29B raise.
- “Built on Bittensor… five to seven times cheaper than Claude Code and Codex at a comparable benchmark.” (08:17)
- Open Incentives: Developers can directly earn tokens for improvements, creating a free market alternative to corporate engineering.
- “Unlike open source where you just get kudos, here you have an incentive layer.” (10:40)
- Liquidity Pathways: Subnet tokens accrue via competition; capable miners (globally, permissionlessly) are strongly incentivized.
4. Investing in Decentralized AI Subnets (17:44 – 26:05)
- Fund Thesis (Stillcore): Invests both in main TAO tokens (the network) and in emerging subnets (specific projects).
- Projects highlighted:
- Ridges: Coding AI copilot
- Targon (Subnet 4): Private, encrypted inference-as-a-service with high-profile investors (Tobi Lütke, Ram Shriram).
- Hippeis: Decentralized, commodity-cheap storage, offering 400–4000x savings over Filecoin.
- Projects highlighted:
- Metrics for Investment: Product competitiveness, team pedigree, tokenomics that tie usage to token value, entry cost.
- Gold Rush Dynamics: Subnets viewed as startups — “we look at these subnets like you would look at startups.” (18:44)
5. The User Experience and Wallets (13:30 – 16:37)
- Guest: Ala Shabana (Crucible Labs/Bittensor co-founder)
- Crucible Wallet: Created for easier allocation of TAO to subnets — “stake and forget.”
- Subnet Listings: Exchanges (e.g. Kraken) exploring listing subnet tokens; complexity due to KYC and regulatory requirements.
6. Mining & Community (24:49 – 26:18)
- Who are the miners? Distributed, global, often pseudonymous contributors (analogous to early Bitcoin).
- Permissionless Opportunity: Any entity with resources can begin mining by following network protocols, lowering the barrier for global participation.
- “There is no onboarding KYC process. It’s permissionless mining, just like a lot of crypto.” (25:02)
7. Project Selection, Quality, and Real-World Adoption (26:23 – 27:41)
- Ala’s View: Key value is in building subnets with “real world” appeal — projects win not just through crypto-native participation, but by solving actual user problems.
- Arms Race: Competition among subnets to secure adoption and tangible use cases.
Memorable Quotes & Moments
-
Jason on Subnet Tokens as Evolution of Open Source Incentives:
“Here you get an incentive layer. … I won a contest to fix something in the product, I get to put that in my wallet…” (11:54) -
Mark on Bittensor’s Market Efficiency:
“Bittensor is the most ferocious form of capitalism and age of excellence that I’ve ever seen.” (31:19) -
Ala on Community & Decentralization:
“Now we’re sort of following the exact same pattern as Bitcoin where … the project is being maintained by really the entire community itself because it is open source and so anyone can contribute.” (36:47) -
Jason summarizing the economic impact:
“We would have a compression of costs that would just be relentlessly being pursued … in a 24/7, 365, billions of tokens pursuit of lowering costs.” (34:47)
Additional Demos & Guest Products
OpenClaw Agent Integrations & Ecosystem (41:47 – 45:45)
- Michael Raya (Setup Claw):
White-glove “OpenClaw in a box” service for founders & execs — SaaS agent setup, proactive automation for busy professionals.
Bisbee – AI-backed Startup Incubator for SMBs (47:52 – 52:49)
- Father/Daughter Team Pete & Maddie Reese:
AI agent-driven business launcher: from ideation to branding to operational team, tailored for new small business founders facing an AI-induced job market shake-up.
Cultural & Entertainment Segment (54:23 – End)
- Bob Dylan’s 1986 AI Prediction (54:23 – 58:49):
- Reflection on art, technology, and soul in music.
- Dylan: "That type of music doesn’t have any roots and it doesn’t have any foundation to it." (55:06)
- Jason and Lon discuss Dylan’s foresight about technology in art and music, referencing documentaries and biographies.
- Lon’s Picks for the Weekend (59:04 – 64:59):
- Civil War miniseries “The Gray House” (Prime Video)
- National Geographic’s “Incas: The Rise and Fall” (Disney/Hulu)
- Werner Herzog’s “Ghost Elephants” documentary (Disney)
- Jason recommends bone-conducting headphones (Shokz) as a practical tech solution for families or outdoor activities.
Timestamps of Key Segments
- 03:56 – Mark Jeffrey joins; intro to Bittensor and programmable mining
- 08:03 – Emergent products on Bittensor subnets (Ridges, Targon, Hippeis)
- 13:30 – Navigating the TAO / subnet token/wallet ecosystem with Ala Shabana
- 18:44 – Investment mindset: subnets as startups, approaches to due diligence
- 24:49 – Profile and dynamics of Bittensor network miners
- 34:47 – AI agents driving cost compression in decentralized infrastructure
- 36:47 – Governance shift: OpenTensor Foundation hands over to community
- 41:47 – OpenClaw agent SaaS for high-leverage professionals (Setup Claw)
- 47:52 – Bisbee AI: founder story, AI-powered SMB accelerator
- 54:23 – Bob Dylan on the “soul” of machine-created art
- 59:04 – Lon’s weekend streaming picks and book/tech recommendations
Tone and Takeaways
- Optimistic, hands-on, and slightly irreverent: The hosts repeatedly celebrate the permissionless, market-driven, and soon-to-be-mainstream nature of decentralized AI. There’s a “gold rush” excitement, but with caution — “20% in things that could go 100x… or go to zero” (40:25).
- Memorable comparisons: Bittensor and its subnets are framed as “startups” competing in a radical “ferocious form of capitalism,” aiming to capture the next phase of the internet’s infrastructure.
- Future-facing: Whether via wallet tooling, agent SaaS, or SMB empowerment platforms, the episode clearly frames decentralized, community-driven AI as a coming wave — with real possibility for both risk and reward.
Notable Quotes
“Bittensor does for stranded talent what Bitcoin did for stranded energy.”
– Mark Jeffrey, 09:52
“Here you get an incentive layer. … I won a contest to fix something in the product, I get to put that in my wallet…”
– Jason Calacanis, 11:54
“The project is being maintained by really the entire community itself because it is open source and so anyone can contribute.”
– Ala Shabana, 36:47
“That type of music doesn’t have any roots and it doesn’t have any foundation to it.” (on machine-generated music)
– Bob Dylan (audio clip), 55:06
Closing Thoughts
The episode paints a vision of a decentralized AI future, where permissionless innovation, open incentives, global talent, and relentless cost efficiency remake both technology and the labor market. Bittensor’s programmable mining rewards, the rise of agent ecosystems, and the melding of blockchain and open source AI are positioned as game-changers. Yet, the need for user-friendliness, real-world application, and “soul” remains vital — echoing both the promise and limits of machine-driven creativity.