Aug 22, 2026 · 1h 14m
Voorhees bets crypto can become AI’s financial infrastructure
Uneasy Money: Why Erik Voorhees Calls AI's Hidden Filter 'Deceptive'
The episode examines whether decentralized finance can give AI agents programmable transactions while preserving privacy and user autonomy.
- 1Venice treats crypto as infrastructure for useful AI products, not merely a self-contained speculative ecosystem.
- 2Its hybrid equity-and-token structure aims to align investors, users, and the platform while exposing tensions around dilution and revenue.
- 3Voorhees argues hidden moderation and centralized control make direct, private AI access a question of individual sovereignty.
Don't miss
Voorhees argues that AI users often interact with models through hidden corporate policies, system prompts, and regulatory constraints, making direct access a sovereignty issue.
The brief
Erik Voorhees frames Bitcoin as proof that technology can create permissionless alternatives when political systems cannot meaningfully control the state.
His broader thesis is that crypto should become financial infrastructure for useful products, especially AI agents that need fast, programmable transactions.
Venice’s hybrid equity-and-token model turns that thesis into a business structure, raising questions about incentives, dilution, buybacks, and alignment.
As an aggregator of closed and open models, Venice emphasizes private, less-censored access while weighing capability, cost, data retention, and hidden policy layers.
Voorhees rejects nationalist AI competition as the central frame, arguing that privacy, sovereignty, and individual freedom matter more than which country leads.
What was said on this episode
26 statements · 14 positive · 12 negative
Meme-coin speculation is consumptive entertainment rather than socially productive activity
“But it's not accretive to society. It is a form of entertainment that is consumptive.”
Listen at 11:39
Utility-token projects faced greater security-classification risk than meme coins
“If, if you had a project that had utility to it and you had a token, it had a higher chance of being deemed a security than if you just released a meme coin.”
Listen at 12:10
Crypto broadly shares principles of decentralization, privacy, and individual sovereignty
“the principles are generally shared. These tend to be things around decentralization. privacy, sovereignty of the individual, decentralization of power.”
Listen at 14:14
AI culture historically favored top-down control of the technology
“the mentality or the philosophy, the ideology within that world was very smart people in academia that believed things should be controlled.”
Listen at 15:00
AI culture previously neglected privacy, association, and user sovereignty
“nobody in AI cared about privacy. Nobody cared about association. Nobody cared about user sovereignty.”
Listen at 16:24
A prevailing AI safety view advocated immediate regulation, surveillance, and censorship
“Government needs to regulate it immediately. Everyone needs to be watched. the prompts that you send need to be censored. The answers need to be censored.”
Listen at 16:37
Centralized frontier AI could make machine intelligence an official state mouthpiece
“that was going to lead to like a very dystopian situation where machine intelligence itself was coming essentially through the official mouthpiece of the state.”
Listen at 17:43
Decentralization tends to localize failures rather than make them catastrophic
“if things are decentralized then at least where the problems happen are localized usually instead of catastrophic”
Listen at 23:47
Crypto is better financial tooling than alternatives
“i actually just think crypto is better financial tooling period full stop”
Listen at 25:21
AI agents will use crypto assets on decentralized rails instead of Wells Fargo accounts
“I don't see a future where like the robots are using bank accounts at Wells Fargo. I do see a future where they're using crypto assets on decentralized rails.”
Listen at 26:19
Gatekeeper-free digital financial technology offers the least-friction value transfer
“natively digital financial technology that does not have a political gatekeeper is is the least friction way for value to move”
Listen at 26:36
VVV token supply will shrink over time
“our entire strategy with VVV is to take it deflationary. So the actual token supply will shrink with time.”
Listen at 38:43
Well-designed tokens can benefit all stakeholder groups
“these are interesting and powerful financial tools that when assembled correctly can be accretive to all groups.”
Listen at 41:58
Anthropic loses substantial money on its $200 monthly plans
“Anthropic is losing money hand over fist on those plus $200 a month plans.”
Listen at 50:19
The cost of AI intelligence has fallen dramatically for several years
“the cost of intelligence is falling dramatically and has been now for a few years.”
Listen at 52:24
AI token sellers will eventually earn no material profits
“The end state is that no one's going to make any material money on selling tokens. That's true of all commodities and tokens are a commodity.”
Listen at 53:17
Chinese AI models deserve preference if they better express individual-freedom principles
“if chinese models better express those principles then i care more about the chinese models than i do about the american models”
Listen at 58:30
The United States is moving toward dystopian control of its population
“america is going down a pretty dark i would say Like a pretty dark dystopian path of trying to control everyone.”
Listen at 58:55
Large-scale data centers will always provide inference more efficiently than local hardware
“From a pure economic perspective, it is always going to be more efficient to just get your inference from large-scale data centers.”
Listen at 1:06:17
Running AI models locally should not be pursued to save money
“the point of running things locally and never and should never be about saving money because you won't.”
Listen at 1:06:26
Server-based models will be roughly ten times more intelligent than edge-run models
“whatever you can run at the edge you can run something 10x is intelligent from a server”
Listen at 1:07:52
State-approved AI servers could restrict users to asking only permitted questions
“if you can only access servers that come through you know like licensed data centers that are approved by the state and will only let you ask certain things, that is the future that I want to avoid.”
Listen at 1:08:03
Corporate and regulatory mediation makes AI interaction opaque, deceptive, and abuse-prone
“your interaction isn't just between like your mind as a human and the machine, which is a beautiful symbiosis. It's got this opaque. undefined, like, MIDI layer, right? And that's not okay. You know, like, that's kind of deceptive and can be abused so easily.”
Listen at 1:09:53
Venice lets users communicate directly with the AI model without an intermediary layer
“in Venice, there is no layer. You just talk directly to the machine.”
Listen at 1:10:29
Venice enables outsiders to attest the entire inference round trip
“The entire round trip of the inference can be attested by an outsider who's mildly technical.”
Listen at 1:11:48
Venice aims to become a mass-market app for unrestricted private AI
“we're trying to build like a mass market. consumer app for, um, for unrestricted and private artificial intelligence.”
Listen at 1:13:17
Statements are attributed to the speaker as said on the episode and reflect their view at the time, not PodLume's. They are not advice.
Weekly email
State of the Pod, every Monday
The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.