
A 25-researcher survey from the Cornell-based IC3 consortium maps crypto-AI convergence, identifies five misconceptions, and says blockchain can aid automation and record-keeping but does not by itself solve AI autonomy, content verification or bias.
The Initiative for CryptoCurrencies and Contracts (IC3), an academic consortium based at Cornell University, published a survey paper on the 8th/Monday examining how crypto and artificial intelligence are converging and where claims about blockchain’s benefits for AI are overstated. Produced with 25 researchers from universities including Cornell, Carnegie Mellon, Princeton, Yale and ETH Zurich, the paper identifies five misconceptions, argues that crypto wallets can help automate AI agents’ transactions without making them truly autonomous, and says blockchain is better suited to timestamping and preserving claims than proving content origin or fixing algorithmic bias. The survey also outlines future research challenges as industry interest in crypto-AI infrastructure, coordination and incentive design grows.