Pearl Network's proof-of-useful-work model faces fundamental headwinds: 85% annual supply growth, dollar-based AI demand, and easily rented GPU capacity undermine token economics.
AI & Agents ·
Pearl Network's proof-of-useful-work model leverages GPU computations from AI training and inference to secure its blockchain, a departure from traditional hash-based mining. However, research circulated in the Blockworks Research newsletter identifies structural vulnerabilities in its token economics: the network faces approximately 85 percent annual supply growth this year, followed by 33 percent the next year, translating to roughly 2.2 million dollars in daily new issuance against an average daily trading volume of 8.5 million dollars. The analysis notes that AI customers pay for inference in US dollars rather than PEARL tokens, eliminating token demand generation from the network's primary use case.
Additional concerns center on the ease of renting GPU capacity. Since inference workloads can run on datacenter GPUs already deployed by the AI industry, the pool of rentable mining hardware substantially exceeds GPUs committed to Pearl mining itself. This surplus capacity mirrors conditions that enabled 51 percent attacks on proof-of-work networks including Ethereum Classic, Bitcoin Gold, and Vertcoin between 2018 and 2020. The research suggests Pearl's daily mining subsidy of approximately 2.2 million dollars lags far behind Bitcoin's daily security budget of approximately 36.6 million dollars, with Pearl's token price needing to appreciate roughly 17-fold to achieve equivalent security economics.
Responses to the critique have contested its framing. One reply characterized the analysis as emotionally toned while highlighting Pearl as the fastest-growing proof-of-work network in history, and another suggested PEARL functions primarily as a collectible asset where the payment method for inference is immaterial. The fundamental dispute around token demand and network security remains unresolved.