Microsoft sees artificial intelligence (AI) as the pivotal technology of our era, actively advancing AI research and investment. However, the tech giant, headquartered in Seattle, is also closely monitoring developments in the world of cryptocurrencies, particularly the potential synergy between blockchain technology and AI.

At the recent Cornell Blockchain Conference in New York, Yorke Rhodes, Microsoft’s director for digital transformation, blockchain, and cloud supply chain, discussed the potential convergence of these technologies. Rhodes suggested that as AI and blockchain progress, it's possible to develop agents that harness the strengths of both. However, he emphasized that we're only beginning to explore this intersection.

During a panel discussion titled "Crypto x AI," Rhodes was questioned by moderator Alex Lin about whether Microsoft plans to create its own blockchain. Rhodes noted that there's already significant innovation in the crypto space, particularly in the open-source community, questioning the need for Microsoft to replicate existing efforts. Instead, Microsoft is focused on enhancing current technologies like layer-2 blockchain rollups, rather than developing its own primary blockchain.

Rhodes acknowledged the growing significance of crypto, stating that it's well-positioned to complement certain types of AI, particularly transformer and diffusion models. However, he recognized that AI tends to dominate discussions and investments, potentially overshadowing emerging technologies like crypto and blockchain.

Lin highlighted the challenges associated with the intersection of blockchain networks and AI, noting the tendency for exaggerated claims and the importance of distinguishing between hype and reality. He pointed out issues such as latency in decentralized networks compared to centralized ones.

Jasper Zhang of Hyperbolic Labs expressed confidence that decentralized networks like blockchains can overcome latency challenges, citing examples where decentralized solutions matched centralized ones in terms of latency.

The conversation also touched upon the trend towards smaller AI models designed to run efficiently on devices like phones and laptops, reducing the reliance on massive computing power. Microsoft has been developing such models, including its Phi-3 family of open models, which approach the capabilities of large language models.

Panelists discussed the likelihood of increased regulatory scrutiny on AI, similar to what cryptocurrencies have faced. They debated the role of regulators in fostering innovation while protecting consumers, acknowledging the challenges of regulating AI due to its opaque decision-making processes.

Lin concluded by asking about the future of AI, including the potential for artificial generalized intelligence (AGI) and advancements in technologies like zero-knowledge proofs (ZK-proofs) and fully homomorphic encryption (FHE).