At The Block’s Emergence conference in Prague on December 10, 2024, leading crypto venture capitalists delivered a sobering assessment of the AI-crypto convergence that has captivated the market throughout the fourth quarter. Rob Hadick, general partner at Dragonfly, and David Pakman, managing director at CoinFund, expressed deep skepticism about whether crypto projects can meaningfully compete with centralized AI giants like OpenAI, Meta, and Microsoft — even as the AI agent token market capitalization has surged past $8 billion.
The Synergy
The theoretical case for AI-crypto convergence remains compelling. Blockchain technology offers decentralized compute networks, verifiable AI training data, transparent model governance, and censorship-resistant AI deployment. Protocols like io.net, which partnered with CreatorBid on December 10 to scale image model training using decentralized GPUs, exemplify the practical applications emerging at this intersection. The DePIN (Decentralized Physical Infrastructure Networks) narrative promises to democratize access to the GPU compute power that AI development demands.
At the market level, the synergy has translated into remarkable price action. AI agent tokens have emerged as the dominant crypto narrative of late 2024, with projects building autonomous AI agents for trading, social media, and decentralized applications capturing significant capital and attention. The sector’s growth reflects genuine demand for AI-native financial infrastructure and automated on-chain operations.
AI Use Cases in Web3
Pakman, however, was quick to ground the discussion in practical reality. He argued that AI’s most meaningful contribution to crypto lies not in tokenized AI projects but in enhancing software development and coding efficiency across the entire ecosystem. AI has significant potential in infrastructure, security auditing, and decentralized application development — what Pakman described as a “massive software and technical revolution” that benefits crypto as a whole rather than creating a separate AI-token category.
The Emergence conference highlighted a growing tension within the crypto industry between builders developing practical AI integration tools and speculators chasing the latest AI-themed token. Projects that leverage AI to improve blockchain infrastructure — from automated smart contract auditing to predictive DeFi risk management — tend to receive more favorable assessments from institutional investors than meme-driven AI agent tokens like Fartcoin and Goatseus Maximus.
Data Privacy Implications
The convergence of AI and crypto also raises critical questions about data privacy and governance. Decentralized AI systems, in theory, could address the concentration of data power among tech giants by enabling federated learning and privacy-preserving model training. However, Hadick’s critique suggests that current implementations are far from achieving this vision. The challenge lies in building decentralized AI systems that can match the performance of centralized alternatives while preserving the privacy and transparency benefits that blockchain provides.
The emergence of AI agents operating autonomously on-chain introduces new privacy considerations. These agents interact with smart contracts, execute trades, and manage assets on behalf of users — creating potential vectors for data exposure if not carefully designed. The industry must develop robust privacy frameworks for AI-crypto interactions before these systems achieve mainstream adoption.
The Innovation Frontier
Despite the skepticism, the innovation frontier continues to expand. Hadick acknowledged one area where crypto and AI agents have clear synergy: payments. “If we have an agent-based future, crypto is an ideal substrate for agent-to-agent and human-to-agent payments,” he noted. This practical application — using blockchain as the payment layer for AI agent interactions — represents perhaps the most viable near-term convergence opportunity.
The io.net and CreatorBid partnership announced the same day illustrates another frontier: decentralized GPU networks enabling smaller AI developers to access compute resources without relying on centralized cloud providers. With over 325,000 verified GPUs on the io.net network, DePIN protocols are building the infrastructure that could eventually support the AI-crypto vision that VCs remain cautious about.
Concluding Thoughts
The Emergence conference revealed a crypto industry grappling with the gap between AI-crypto hype and practical reality. While the $8 billion AI agent token market reflects genuine market interest, the skepticism from experienced venture capitalists suggests that the current cycle’s AI tokens may face a reckoning similar to the 2021 gaming token boom. The sustainable winners in the AI-crypto space will likely be projects that solve real infrastructure problems rather than those riding narrative momentum. For investors and builders alike, the message from Prague is clear: focus on utility over speculation, and recognize that the most impactful AI-crypto convergence may happen quietly, in the infrastructure layer, rather than loudly in the token market.
Disclaimer: This article is for informational purposes only and does not constitute financial advice. Always conduct your own research before making investment decisions.
$8 billion market cap on AI agent tokens and not a single one has a working product that competes with OpenAI. Dragonfly and CoinFund are right to be skeptical here.
to be fair, OpenAI had no competing product 3 years ago either. the question is whether crypto can ship fast enough to matter
Rob Hadick from Dragonfly basically said what everyone in the space thinks but wont say out loud. centralized AI is years ahead and crypto cant just wish that away
Rob Hadick isnt saying crypto cant compete. hes saying the current batch of AI tokens are premature. theres a difference
io.net partnering with CreatorBid on decentralized GPU training is the kind of thing that makes the 8B market cap defensible. actual compute happening on actual distributed hardware, not just tokenized buzzwords
8B market cap for AI agent tokens and not one protocol can show meaningful revenue. Hadick was right to push back
$8B in AI tokens and VCs are still saying “where’s the real revenue?”
Dragonfly and CoinFund calling out the emperor has no clothes on AI-crypto convergence
Dragonfly and CoinFund are invested in DePIN plays tho. their skepticism is strategic, not altruistic. follow the money
vc_trackr_ of course they are. publicly skeptical while their portfolios are loaded with DePIN and AI tokens. classic VC playbook, talk down the market while accumulating