While everyone is debating whether ChatGPT is making students cheat, a more insidious liquidity crisis is unfolding—one that drains the value of authentic human output. Author Dave Eggers recently warned OpenAI employees that their creation is causing 'catastrophic impact' on education. He’s not wrong. But he’s only seeing half the balance sheet.
The warning, delivered directly to the engineers building the models, carries weight. Eggers is a novelist who understands narrative, credibility, and the economics of attention. His concern centers on how generative AI destroys the intrinsic value of original student work. Yet the mainstream conversation—banning ChatGPT in classrooms, deploying detection tools—misses the deeper structural shift. This is not a problem of cheating. It is a problem of provenance. And proven once is a liquidity problem.
Context: Dave Eggers, author of The Circle and a long-time advocate for privacy and authentic human expression, spoke to OpenAI staff in a closed-door meeting. His message was blunt: ChatGPT is having a catastrophic impact on education. He cited the erosion of critical thinking, the collapse of writing as a skill, and the impossible challenge of assessing student work when AI can produce indistinguishable essays in seconds. The article in Crypto Briefing also flagged that this warning raises questions about 'crypto identity'—a term that suggests a possible solution rooted in blockchain technology.
Education is the current canary in the coal mine, but the issue is universal. Every industry that relies on verified human contribution—journalism, legal documents, medical records—faces the same threat. Generative AI floods the market with near-perfect substitutes, crashing the price of authentic work. This is a liquidity crisis in intellectual capital.
Core: The provenance liquidity trap
Let me explain using a framework I developed during the DeFi Summer of 2020. In DeFi, liquidity fragmentation occurs when trading volume is spread across multiple pools, making it impossible to determine the true price of an asset. Fake volume from wash trading misleads allocators. The same happens now with text output. A student writes an essay—real labor, unique insights. Another student prompts ChatGPT to generate a paper in 30 seconds. Both produce similar outputs. The second one appears to have identical value, but it carries no intellectual cost. The market—in this case, the grade curve—cannot differentiate. The real essay gets devalued. This is not academic dishonesty; this is a systemic failure of verification.
Watch the flow, ignore the noise. The flow here is the supply of digital content. Generative AI has collapsed the marginal cost of producing written output to near zero. When supply becomes infinite, price crashes. The only way to restore value is to introduce artificial scarcity—through cryptographic proof of human origin. This is why crypto identity matters.
My experience during the ICO bubble taught me to spot value illusions. In 2017, 80% of projects had no sustainable tokenomics; they relied on liquidity inflows. Similarly, today, the 'value' of an AI-generated essay is a phantom—it exists only until someone demands proof of its origin. The parallel is exact. Just as I liquidated 70% of my positions before the regulatory crackdown, smart allocators now will short the commoditized content market and long the infrastructure that verifies human work.
But most analysts are looking in the wrong direction. They focus on banning AI or building detection tools that will be obsolete within months. That’s like trying to plug a leaky dam with bubble gum. The fundamental flaw is not the tool; it’s the lack of an authentication layer at the network level.
Contrarian: The decoupling thesis
Here comes the counter-intuitive take. While the crowd panics about AI replacing teachers, the real decoupling is that AI will actually increase the demand for certified human work. Let me explain through a financial lens. In a bull market, every token rises together. But eventually, quality decouples from noise. The same will happen in education. Once employers and universities realize that GPT-generated content is indistinguishable from human output, they will place a massive premium on provably human work. This is not a prediction of doom; it’s a prediction of bifurcation.
Crypto identity—a blockchain-based record of human authorship, verified through proof-of-personhood mechanisms—becomes the new credential. Forget diplomas. The signal will shift to cryptographic signatures tied to biometrics or social verification. This shift is already being built. Protocols like Worldcoin, ENS with attestations, and decentralized identity solutions are positioning for this moment. DeFi yields are traps, not gifts—this one is genuine.
The trap is the assumption that AI detection software will solve the problem. It won’t. Detection is an arms race that the attacker will always win because generative models can be fine-tuned to evade classifiers. The only sustainable solution is to change the verification layer from 'detect after creation' to 'prove at creation'. This is a cryptographic problem, not a machine learning problem.
AI-generated content is digital vanity metrics—they look impressive but carry no fundamental value. The real value lies in the protocol that timestamps and signs human creation. Think of it as a proof-of-work for humans.
Takeaway: Position for the cycle of certification
For institutional allocators watching this space, the play is clear. The next 12-24 months will see a wave of regulatory activity around digital identity, especially in the EU with eIDAS 2.0 and its blockchain provisions. Simultaneously, education systems will be forced to adopt some form of proof-of-human-work to maintain credibility. The intersection of these two trends creates a massive opportunity for protocols that offer verifiable, sovereign identity.
Ignore the noise about AI replacing teachers. That’s the easy narrative. The real story is the liquidity crisis in human output and the infrastructure needed to restore scarcity. The alpha is in identity protocols that bridge Web3 and educational institutions. Watch the flow of regulatory clarity; ignore the hype around new LLMs.
I have structured our fund’s macro position accordingly: long on decentralized identity infrastructure, short on any platform that aggregates content without provenance verification. The collapse of the Terra-Luna ecosystem taught me that without real collateral, everything is a mirage. The same applies here: without cryptographic proof of human origin, all digital content is subject to a run on authenticity.
The bubble of generative AI will not pop—it will deflate into two distinct markets: verified human work trading at a premium, and unverified synthetic content trading at zero. Crypto identity is the only real hedge against this future. Watch the flow, ignore the noise.