{"id":503387,"date":"2025-11-02T19:00:41","date_gmt":"2025-11-02T19:00:41","guid":{"rendered":"https:\/\/cryptoslate.com\/?p=503387"},"modified":"2025-11-01T13:29:12","modified_gmt":"2025-11-01T13:29:12","slug":"can-blockchain-tame-ais-ip-problem","status":"publish","type":"post","link":"https:\/\/cryptoslate.com\/can-blockchain-tame-ais-ip-problem\/","title":{"rendered":"Can blockchain tame AI’s IP problem?"},"content":{"rendered":"

The following is a guest post and opinion from Shane Neagle, Editor In Chief from The Tokenist.<\/em><\/p>\n

It is no secret that large language models (LLMs) crossed the capability threshold by harvesting vast amounts of public and private data. Combined with breakthroughs in transformer architectures and compute power, this data scraping led to concerns about intellectual property (IP) rights.<\/span><\/p>\n

Intellectual property frameworks exist to incentivize innovation and creative spark, protecting creators and businesses. In turn, the entire society benefits from that incentive structure. Eventually, IP protections typically expire, at which point IP becomes integrated into the public domain.<\/span><\/p>\n

The global harmonizing IP framework is the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS Agreement) under the World Trade Organization (WTO) umbrella, together with the World Intellectual Property Organisation (WIPO).<\/span><\/p>\n

However, with AI rapidly blurring the line between human and machine creativity, the foundational assumptions of the IP system are under strain. Without explicit consent and compensation, LLMs are routinely trained on copyrighted works, eroding the important incentive structure.<\/span><\/p>\n

\"The
The AI Model Development Cycle<\/figcaption><\/figure>\n

Over time, it is not difficult to see digital oligopolies entrenching their power with the largest compute power and data access, while barring smaller players from large-scale data scraping.<\/p>\n

Yet, once again when it comes to data flows, a potential solution can arise from the blockchain ecosystem. Specifically, with layer-1 Camp Network (CAMP) blockchain.<\/p>\n

How Does Camp Network Tackle IP Incentivization Erosion?<\/h2>\n

Just as Bitcoin mainnet immutably registers the transfer of value, Camp Network aims to immutably register the transfer and attribution of people\u2019s work. With a permanent and verifiable record of ownership, creators can automatically enforce licensing terms – through smart contracts – whenever AI models use this registered content.<\/p>\n

To accomplish this, Camp Network uses the proof-of-provenance (PoP) protocol, which handles IP origin and licensing terms. On top of this core protocol, Camp Network uses BaseCAMP as a global IP registry and SideCAMPs tailored for dApps pertaining to IP enforcement in various sectors such as music, books, or gaming.<\/p>\n

Specifically, creators and organizations could register IP as non-fungible tokens (NFTs), which have already pioneered a form of on-chain royalties system, despite the market deflation since late 2021. Although this was an important proof of concept, royalties enforcement relies more on marketplace cooperation, not the blockchain itself.<\/p>\n

Case in point, there was a period of \u2018royalty wars\u2019 between OpenSea and other marketplaces like Blur, in which marketplaces introduced a royalties-optional model opting for token incentivization instead. Eventually, this led to a fragmented and voluntary NFT royalty ecosystem, exacerbated by faulty metadata<\/span><\/a>.<\/p>\n

Building on these lessons, Camp Network is embedding royalty logic at the protocol layer, instead of relying on marketplaces. This means that royalties would not just apply to content, but to data usage – ideal for scenarios when LLMs train on registered datasets.<\/p>\n

Camp Network\u2019s Registration and Monetization Process<\/h2>\n

With its purpose-built layer-1 blockchain, Camp Network is foremost aimed at individual creators across all digital content categories. Whether it is a music track or a digital image, the process is as follows:<\/p>\n