Burian says crypto’s lessons also apply to AI investing
According to ChainCatcher, Blockchain Capital investor Jonah Burian said crypto has sped up market cycles, while also giving early-stage companies liquidity through tokens. He added that market behavior in crypto is public by default, making the sector’s lessons relevant to AI investing as well.
Burian wrote that large outcomes are contagious and tend to trigger FOMO. After Bitcoin became a trillion-dollar asset, Ethereum and Solana showed that more massive outcomes were possible, which helped fuel the rise of alt-L1 trades. Venture capital firms then backed new Layer 1 projects as if they were lottery tickets.
He compared today’s AI lab financing to the alt-L1 era
Burian said a similar pattern is now playing out in AI. OpenAI and Anthropic are moving toward trillion-dollar scale, and each new lab is being priced with the same lottery-ticket logic.
He noted that some Layer 1 projects once raised capital at multi-billion-dollar valuations on the strength of a white paper and founding team alone. Now, new AI labs are raising at multi-billion-dollar valuations with research papers and teams recruited from OpenAI, Anthropic, or Google DeepMind.
AI pricing is forming in less transparent markets
He also said that once hot money enters a sector, fast-moving speculative capital usually follows. Crypto has already gone through phases such as the token-as-product trade and structures built around high FDV and low float. In his view, similar behavior is now appearing in AI.
Still, he pointed to one major difference: AI pricing is being formed in opaque, semi-liquid secondary markets, unlike crypto tokens, which trade in public markets.
Value may shift away from model providers
Burian added that blockspace moved from scarcity to abundance and eventually became a commodity, with applications capturing most of the value. He argued AI could follow a similar path, where models become commoditized and value moves both up and down the stack, leaving applications and hardware to capture profits while the model layer faces pressure.
He said periods of market mania often lead financial capital to overfund infrastructure. On that basis, he argued that current bets on new AI labs only make sense if returns to large-scale R&D actually hold. Crypto and alt-L1 history, he wrote, should be enough to justify skepticism, unless AGI arrives.