Grayscale head of research Zach Pandl highlighted three AI-focused crypto projects in an October 23 note, naming DATA Network, USD.AI, and Pearl as assets the firm is watching amid the growing convergence of artificial intelligence and cryptocurrency.
DATA Network collects human-contributed training material such as photos, video, and voice recordings. Applications on the network license datasets to AI labs and reward contributors. The Sui Foundation announced on October 12 that DATA's Trace registry, launched in June, has registered more than 250 million contributions and processes over 3 million receipts daily. Apps including Kled, Numo, and Oto license data to frontier AI labs and robotics companies. However, published figures do not disclose how many contributions were purchased or how much contributors received. The Sui Foundation is investing $30 million in DATA Network, becoming its largest institutional holder, as DATA migrates to Sui and Walrus. Its own chain will be retired, and DATA tokens will carry over one-for-one, with governance and holder economics described as at least equivalent to current terms. A transition plan is due before the upgrade, but neither announcement explains how licence revenue reaches DATA holders.
USD.AI / CHIP is described by Grayscale as arranging loans backed by GPU equipment, with on-chain pools that let participants earn yield. Borrowers are compute operators needing capital before their GPUs pay for themselves. USD.AI's documentation separates the mechanics: USDai is a stablecoin backed one-to-one by PYUSD with no yield, while sUSDai earns interest linked to GPU loans and reserves. CHIP is the governance token and, according to its MiCA white paper, confers no equity, dividend, or profit rights. Staking CHIP in insurance modules is optional, and a possible fee switch for tokenholders is listed as a future protocol decision. USD.AI says an updated white paper is being prepared.
Pearl has risen around 580% in three months, according to Grayscale citing CoinGecko data as of October 22. Pearl aims to let GPUs run AI workloads and secure a proof-of-work blockchain through matrix multiplication, earning PRL rewards on top of AI revenue. Together AI launched its first Pearl-powered product, a discounted inference endpoint, on May 15, with a discount of more than 25% offset by the future value of crypto emissions. A June preprint by researcher Abhinaba Basu challenged the useful-work claim, concluding the network "produces zero useful AI computation," though the study is single-author, not peer reviewed, and measured the network before the three-month rally Grayscale cites.
Grayscale noted that long-term relevance will depend on adoption and competitive economics, and its disclaimer states the firm and affiliates may have interests in the assets discussed.
