Editing Openai/68de3c2d-a7fc-800f-bc1a-9b1928a73b37 (section)

==== Key Investable Categories in Crypto × AI ====

Major investor theses converge on several high-potential categories at the AI/blockchain intersection. These categories, summarized in Table 1 below, include agentic protocols (for autonomous AI agents), AI-optimized blockchains, decentralized compute networks, and tokenized AI models/data. Each category addresses specific needs at the nexus of AI and crypto – from enabling machine agents to natively participate in on-chain economies to providing the decentralized infrastructure (compute, data, identity) that AI systems can leverage.

Table 1: Emerging Investable Categories at the Crypto–AI Intersection

| Category                       | Description / Thesis                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     | Recent Examples & Backers                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                            |
| ------------------------------ | ------------------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------ |
| Agentic Protocols & AI Agents  | Protocols and tools for AI agents to transact and communicate on-chain autonomously. These enable “agent-to-agent” commerce with minimal human input
a16zcrypto.com
americanbanker.com
. Blockchains provide open standards for identity, payments, and interactions between agents, allowing AI systems to coordinate and execute tasks on their own
a16zcrypto.com
.                                                                                                                                                                                                                                                                   | x402 Protocol – A standard for machine payments (launched by Coinbase & Cloudflare) to let AI agents use HTTP 402 for automated payments
americanbanker.com
.
Circuit & Chisel – Startup developing an agent-to-agent payment standard (ATXP); raised $19 M from investors like ParaFi, Coinbase Ventures, Solana Ventures, etc
americanbanker.com
.
Halliday, Catena, Skyfire – Examples of projects building on-chain rails for agent interactions (payments, workflows)
a16zcrypto.com
.                                                                                                                                                                                                                                                                                                                                                                          |
| AI-Native Blockchains          | New layer-1 blockchains or app-chains optimized for AI workloads and agent economies. Investors note a “fat app chain” renaissance – the idea that major AI applications might run on dedicated chains for speed and customizability
bitget.com
. These AI-native blockchains integrate AI logic or support high-throughput microtransactions that AI agents require.                                                                                                                                                                                                                                                                    | 0G (Zero Gravity) Aristotle – An AI-focused L1 blockchain; among the first “AI-native” chains, it launched a mainnet and token for on-chain AI logic
u.today
.
“Unichain” Thesis – Pantera Capital’s thesis that each significant AI application could have its own chain (reviving the fat application chain model) to handle its unique demands
bitget.com
.
Avalanche Cogitus – Example initiative to help launch AI-specific subnets/L1s (announced by Ava Labs), indicating interest in AI-tailored blockchain infrastructure.                                                                                                                                                                                                                                                                                                                                  |
| Decentralized Compute & Data   | Distributed computing networks and data marketplaces to support AI training and inference. Crypto tokens incentivize participants to contribute idle GPU/TPU resources or datasets, democratizing access to AI compute
a16zcrypto.com
variant.fund
. This lowers costs and reduces reliance on Big Tech clouds
a16zcrypto.com
a16zcrypto.com
. Related are data DAO models that reward users for supplying high-quality data for model training
variant.fund
.                                                                                                                                                                           | Gensyn – Decentralized GPU compute marketplace for AI model training; backed by a16z (cited by a16z as a leader in distributed ML training/fine-tuning)
a16zcrypto.com
.
Render Network (RNDR) – A community GPU network (initially for rendering) now applied to AI workloads, illustrating tokenized compute provisioning.
Ocean Protocol – A data marketplace protocol where AI developers can access or monetize datasets via tokens (an early example of on-chain data exchange).
Worldcoin – Proof-of-personhood project (backed by Variant) that, while controversial, amasses a unique biometric dataset to enable human ID in an AI-driven internet
variant.fund
.                                                                                                                                                                                          |
| Tokenized AI Models & Services | AI models, services, or content packaged into tokens or DAO structures for community ownership and monetization. The theses predict a world of “millions of models”, where crypto provides the substrate for funding, governing, and distributing these models
members.delphidigital.io
. By tying AI models or APIs to tokens/NFTs, developers can raise capital and users can invest in or govern AI development
cyber.fund
. Crypto also enables micro-payments for AI services (e.g. paying per API call or model query on-chain) and marketplaces for AI-generated content with provable provenance
a16zcrypto.com
a16zcrypto.com
. | Botto – An AI art DAO (decentralized autonomous organization) where a generative art model’s outputs are curated by a community; the $BOTTO token lets holders guide the AI’s creative direction (Variant-backed)
variant.fund
.
Sastra ($TIA) – The Celestia modular blockchain is being leveraged to host AI models on-chain (Placeholder’s Chris Burniske has discussed $TIA as enabling app-specific chains for use cases like AI).
Onchain AI Models – Vision outlined by funds like cyber•Fund: future AI models deployed on-chain with weights and usage controlled via smart contracts; this makes the model a “commoditized” asset investors can directly own or fund
cyber.fund
. For instance, rights to an AI model’s output could be tokenized, or fine-tuned model instances launched via token sales (as seen with recent on-chain agent launchpads). |

Table 1: Key categories in Crypto × AI investing, with descriptions and examples. These span infrastructure (compute, blockchains), middleware (protocols for agents), and applications (tokenized models, data, and services). Investors see these areas as fundamental building blocks of an “AI-powered, blockchain-governed” interneta16zcrypto.com<ref>{{cite web|title=a16zcrypto.com|url=https://a16zcrypto.com/posts/article/ai-crypto-crossovers/#:~:text=The%20economics%20of%20the%20internet,or%20broad%20communities%20of%20users|publisher=a16zcrypto.com|access-date=2025-11-25}}</ref>a16zcrypto.com<ref>{{cite web|title=a16zcrypto.com|url=https://a16zcrypto.com/posts/article/ai-crypto-crossovers/#:~:text=%E2%80%94%20but%20it%E2%80%99s%20often%20been,the%20AI%20of%20the%20future|publisher=a16zcrypto.com|access-date=2025-11-25}}</ref>.

Each category reflects a synergy between AI capabilities and crypto networks. For example, the push for agentic protocols stems from the rise of autonomous AI agents that can act as economic actors. Crypto provides these agents with identities, wallets, and payment rails to operate independently – something not possible with traditional finance APIsbitget.com<ref>{{cite web|title=bitget.com|url=https://www.bitget.com/news/detail/12560604296111#:~:text=optimal%20strategies%20based%20on%20market,conditions%20and%20user%20preferences|publisher=bitget.com|access-date=2025-11-25}}</ref>. The user’s thesis similarly emphasizes AI agents having their own wallets and a unified agent communication protocol (e.g. referencing the x402 initiative), an area receiving heavy investment now. Likewise, the concept of AI-specialized blockchains shows up both in venture theses and the user’s focus on “AI-native” on-chain ecosystems. Across the board, these categories aim to decentralize the AI value chain – from computing power to data, models, and agent interactions – using crypto-economic incentives.