From chasing the next star public blockchain to betting on the next OpenAI, investors' expectations feel familiar, though the returns may not be as hoped.

By: @jonah_b, Researcher at Blockchain Capital

Translated by: Jiahua, ChainCatcher

It is hard not to notice the striking similarities between the current AI frenzy and previous waves of enthusiasm in the cryptocurrency market.

The crypto market serves as a great observation point for how people act in the face of significant technological waves: it has its positives, its negatives, and its most dreadful aspects. This is due to the rapid pace of cycles in the crypto market, where early projects can achieve liquidity through tokens that other markets struggle to provide, and market behavior is, by default, public as blockchain data is open.

We have learned a lot from this market, and these lessons are applicable to AI as well. If you are investing in AI, this article is for you.

Lottery-like Bets

Massive successes trigger follow-the-leader behavior and create FOMO. Once a category of assets births a groundbreaking winner, investors flock in, attempting to replicate its success path. However, second-wave projects rarely reach the heights of the first wave, and many projects ultimately become castles in the air, leading vast sums of money to go down the drain.

Bitcoin became a trillion-dollar asset. Subsequently, Ethereum turned into a hundred-billion-dollar asset, and Solana also became a multi-billion dollar asset, proving that this market can produce more than one massive winner.

Thus, the wave of investment in new public blockchains arose. Venture capital firms treated these projects like lottery tickets; hitting just one could yield multiples of the entire fund size.

Today, the valuations of many emerging AI labs are also based on lottery-like expectations.

OpenAI and Anthropic are both approaching trillion-dollar valuations. The formula behind this is simple: first, there is an extremely popular market, then comes a proven successful follower. Hence, every new entrant is viewed as the next lottery ticket to wealth.

In the past, many Layer 1 projects received multi-billion dollar valuations almost solely based on a white paper and a founding team. The stories they told were also very enticing:

What if the global economy runs on our chain?

Similarly, emerging AI labs have raised tens of billions based on a set of research ideas and founder teams poached from OpenAI, Anthropic, or GoogleDeepMind.

What if they actually manage to create the "God of Machines"?

However, many times, this investment merely bets on the appearance of the next higher-priced buyer.

Many early investors may not be genuinely assessing the current fundamentals and future prospects of the company. They know that the arrival of a star talent or a partnership with a hyper-scale cloud vendor can attract investors to hike the company's valuation once more. Coupled with the increasing liquidity in secondary markets today, they assume that the next buyer will always appear.

Market Chaos

As BCAP GP @CremeDeLaCrypto mentioned on the Bankless show, when large amounts of capital flow into the market, there will always be "a group of speculators and scammers chasing quick money, rushing wherever it's hot."

For over a decade, crypto investors have been witnessing such phenomena firsthand: batch after batch of projects, parading under the banner of "tokens are the product," pushed to market through dubious market-making techniques, investor-unfriendly high FDV/low circulating supply structures, unfavorable SAFT agreements, and an endless array of other market tricks.

Now, the AI industry is playing out a similar scene. For instance, those structures made up of three-layer SPVs, charging exorbitant fees...

However, there is a key difference between the two: in the crypto market, prices are public, and tokens can be traded openly. In AI companies, prices and valuations are formed in opaque, relatively illiquid secondary markets.

Even without discussing speculative behavior, AI investors can see from the crypto market that changes in market structure will determine where profits ultimately flow.

For example, the significant investments that AI investors are making now may end up becoming standardized commodities.

Case from the Crypto Market: Block Space Becomes a Commodity

Block space used to be scarce and expensive, hence capital poured in to try to expand supply.

However, the industry later went a bit too far: more and more Layer 1 chains launched, and Ethereum also increased Layer 2. Ultimately, block space shifted from scarcity to abundance, even over-saturation.

This is good for technological development, but not necessarily so for investors. Today, many alternative Layer 1 revenues remain very limited.

The early market generally bet on the "fat protocol" theory, but the "fat application" theory ultimately gained the upper hand.

As block space became cheaper, users spent less on underlying infrastructure, while spending on upper-layer applications increased. Application layer projects like Tether, Hyperliquid, Aave, and Polymarket ended up capturing the bulk of the revenue.

Case of AI: Models May Become Commoditized

AI may be replaying a scene that the crypto market has experienced early on.

Chinese AI labs are continuously making increasingly powerful model weights public. I have explained the incentive mechanisms behind their actions.

If model weights become standardized commodities, model prices will continue to fall, transferring value to the upstream and downstream of the technology stack.

Applications will be the primary beneficiaries: if the marginal cost of using models approaches the marginal cost of running models, applications will no longer have to pay for the high profits at the model layer.

OpenAI and Anthropic may not be directly impacted by this change, as they already have user bases, enterprise customer relationships, and distribution channels aimed at developers. But they are exceptions.

In this sense, they are more like Hyperliquid in the AI field, rather than merely providing underlying infrastructure like Layer 1.

In other words, OpenAI and Anthropic have achieved vertical integration. Other AI labs unable to directly reach end-users may find their situation more challenging.

Another result of the declining model profit margins is that energy and hardware at the bottom layer of the technology stack have the opportunity to gain higher profit margins. This trend may be even more evident when computational power supply is constrained by physical conditions.

In short, the hardware layer and application layer may capture more profits, while the model layer will be squeezed. Yet considerable investment funds are flowing precisely towards the model layer.

This is Not New

From a larger cyclical perspective, this is actually not surprising.

Carloota Perez, in "Technological Revolutions and Financial Capital," argues that these types of technological revolutions go through several stages: infrastructure laying, frenzy, collapse, and deployment.

Financial capital excessively invests in infrastructure during the frenzy, but this overbuilt, now inexpensive infrastructure will ironically support the development of the next generation of applications.

This helps explain why there was excessive construction of block space in the crypto market. The expansion of models in the AI field may become the next case.

Investing in emerging AI labs necessitates believing that they can yield enormous returns on R&D investments. The history of the crypto market and alternative Layer 1 should at least make us question this assumption.

Of course, if AGI emerges, the above judgments may not hold. Because we have no idea what the economy will look like after the arrival of AGI.