The Case for Congressional Focus on Decentralized AI

As Congress conducts hearings with SEC Chairman Gary Gensler and pushes to better regulate and stimulate the evolving digital economy, it must recognize the unique needs of decentralized AI – a critical yet often overlooked sector at the intersection of blockchain and artificial intelligence.

Despite this crossover, decentralized AI cannot be legislated through a financial lens, nor can it be forced into AI regulations. Given its overlap in distinct sectors, however, there’s a real chance lawmakers will try to fold it into AI and crypto bills – or overlook it altogether – which would be a missed opportunity for innovation in this country.

Powerful framework

Simply put, decentralized AI allows for the distribution of data, computation, and decision-making processes across multiple devices or nodes, enabling them to work together without relying on a centralized authority, often utilizing open source software and models. This gives developers the tools to share their data collaboratively to build AI models, and to access compute from a diverse range of sources. It’s a powerful framework that empowers these developers to contribute to the AI ecosystem without the need to manage the entire process themselves, enabling researchers and startups to participate in a field where rising costs and difficulty of access threaten to push them out.

And while Congress is making welcome progress to advance comprehensive legislation that would clarify the rules of the road for crypto, the bills are one-dimensional – financial in nature – and don’t address the obvious difference between an underlying digital asset of a protocol versus the decentralized AI applications running on the same protocol. Financial regulators shouldn’t end up overseeing decentralized AI just because projects issue tokens; that would be like the SEC regulating toothpaste because Johnson & Johnson issues common stock.

The next wave

It’s important we get this right, because decentralized AI is a critical field that can salvage the organizations that gave birth to the AI revolution – and possibly bring us the next wave. The universities that created the concept of machine learning and neural networks are under threat, as they cannot compete with Big Tech as it snaps up the world’s GPUs and top talent. In a similar vein, the kinds of startups that brought AI to market are facing similar constraints and oftentimes must defer projects. Without those engines of U.S. innovation working properly, progress will rest in the hands of a few large corporations. It’s a lesson that lawmakers will want to heed, as these universities and small businesses help form the fabric of the American economy, and create jobs and opportunities for constituents.

Congress’s turn

All this goes to show the importance of establishing rules and regulations that allow decentralized AI to flourish. Congress must establish clear regulations for data privacy and security to protect individuals’ information when processed locally, and promote transparency and accountability in AI algorithms that can protect ethical standards and prevent bias in decentralized systems.

Lawmakers should consider a unique regulatory framework for decentralized AI, separate from traditional AI regulations in the Algorithmic Accountability Act and from financial guidelines covered in the Blockchain Regulatory Certainty Act. The new legislation would need to address the particular challenges of decentralized AI, ensuring transparency in algorithmic decisions, and providing guidelines on accountability in decentralized networks. Further, Congress should look at expanding funding opportunities under the National AI Initiative Act to ensure decentralized AI projects – especially those from universities and small businesses – receive government R&D funds and tax incentives, preventing Big Tech from monopolizing AI innovation. Along those lines, Congress would be wise to consider bolstering antitrust enforcement to stop major corporations from controlling critical AI infrastructure.

Note: The views expressed in this column are those of the author and do not necessarily reflect those of CoinDesk, Inc. or its owners and affiliates.