Guest: Don Gossen, CEO at Nevermined
Host: Seth Earley, CEO at Earley Information Science
Chris Featherstone, Sr. Director of AI/Data Product/Program Management at Salesforce
Published on: January 13, 2025
In this episode, hosts Seth Earley and Chris Featherstone are joined by Don Gossen to explore the transformative potential of AI and blockchain in decentralized energy networks. This episode explores how AI and blockchain technologies work together to optimize energy consumption, secure transactions, and drive innovation in local energy management. Don shares real-world case studies showcasing the impact of AI-driven IoT solutions, from reducing costs to enabling energy sovereignty in European markets.
Tune in to hear Don’s insights on the cutting edge of AI, blockchain, and the future of decentralized energy.
Quote from the show:
"Blockchain technology offers elegant solutions to problems of provenance in data and ML models, enabling higher fidelity and trustworthiness. The integration of decentralized blockchain with centralized ML technology presents conflicts, but it is essential for developing better solutions and new economic models. It's about more than just replacing tasks; it's about redefining how we manage and transact value in an increasingly autonomous ecosystem." – Don Gossen
Links:
LinkedIn: https://www.linkedin.com/in/donald-gossen-40ab96/
Website: https://nevermined.io
Ways to Tune In:
Earley AI Podcast: https://www.earley.com/earley-ai-podcast-home
Apple Podcast: https://podcasts.apple.com/podcast/id1586654770
Spotify: https://open.spotify.com/show/5nkcZvVYjHHj6wtBABqLbE?si=73cd5d5fc89f4781
iHeart Radio: https://www.iheart.com/podcast/269-earley-ai-podcast-87108370/
Stitcher: https://www.stitcher.com/show/earley-ai-podcast
Amazon Music: https://music.amazon.com/podcasts/18524b67-09cf-433f-82db-07b6213ad3ba/earley-ai-podcast
Buzzsprout: https://earleyai.buzzsprout.com/
Podcast Transcript: AI Agents, Blockchain, and the Future of Decentralized Energy Networks
Transcript introduction
This transcript captures a conversation between Seth Earley, Chris Featherstone, and Don Gossen about integrating AI and blockchain technologies for decentralized energy networks, exploring data provenance, autonomous AI agents, real-world energy market applications, and the infrastructure needed for AI agents to participate as independent economic actors.
Transcript
Seth Earley: Welcome to the Earley AI Podcast. My name is Seth Earley and I'm Chris Featherstone. Really excited to introduce our guest for today to discuss some of the misconceptions around AI, some of the ways that we can think about AI and Web 3.0, how blockchain and crypto technologies fit in to getting the greatest value from AI and from machine learning and from agents and really talking about some of the challenges that are facing implementations of AI and how assigning value to those interactions and those transactions can really start to release the real potential of AI and agents and really incentivize exchange of value and exchange of information across the ecosystem and really to start to turbocharge how people are implementing and getting value from AI. So our guest today is a seasoned technologist. He has over 20 years of experience in machine learning and IT consulting. He's worked for some of the largest companies in the world, like HSBC, L'Oréal, AXA. Not only has he founded cutting edge, co-founded cutting edge AI projects such as Ocean Protocol, but he's now at the helm of Nevermined, a company revolutionizing the way AI agents interact with financial systems. Don Gossen, welcome to the show.
Don Gossen: Thanks for having me and Happy New Year.
Seth Earley: Happy New Year to you. So Don, let's start with your background. Tell us about your journey and what led you to focus on the intersection of AI and blockchain.
Don Gossen: Sure. I've been working in machine learning for over 20 years now. I started in traditional enterprise IT, working with large corporations on AI implementations. Over time, I became increasingly interested in the infrastructure challenges around AI—not just building models, but creating systems where AI can actually be deployed reliably and securely at scale. That's what led me to blockchain. I saw that blockchain offered elegant solutions to some fundamental problems around data provenance, model trustworthiness, and creating economic systems where AI agents could participate autonomously.
Seth Earley: When you say provenance, what exactly do you mean? Why is that important for AI?
Don Gossen: Provenance means knowing where something came from and its complete history. For AI, this is critical. If you're using a machine learning model to make decisions, you need to know: What data was it trained on? Who created that data? Has it been tampered with? What's the lineage of the model itself? In traditional systems, this information is often lost or can be manipulated. Blockchain creates an immutable record of all of this. Every data source, every training run, every model version—it's all recorded on chain. That creates trust and enables auditing in ways that weren't possible before.
Chris Featherstone: So this is about creating a chain of custody for AI systems?
Don Gossen: Exactly. Just like in legal or scientific contexts, you need chain of custody. With AI, especially as it becomes more autonomous and makes more consequential decisions, you need to be able to trace back every component. Blockchain provides that infrastructure.
Seth Earley: Let's talk about AI agents. You mentioned that they need to interact with financial systems. What does that mean in practice?
Don Gossen: Think about it this way. Right now, when we talk about AI agents, we usually mean software that acts on behalf of a human. The human pays for things, the human owns things, the human makes decisions. But as AI agents become more sophisticated and more autonomous, they need to be able to participate in economic systems directly. If an AI agent is going to interact with the world economically, it needs to be able to pay for things. It needs a wallet. It needs to be able to transact without requiring a human to approve every single micro-transaction.
This is where blockchain becomes essential. You can give an AI agent a wallet on a blockchain. It can hold cryptocurrency or tokens. It can make payments. It can receive payments. It can enter into contracts. All of this happens programmatically, securely, and with full auditability. That's what we're building at Nevermined—the infrastructure for AI agents to participate in economic systems autonomously.
Chris Featherstone: Can you give us a concrete example of where this is being used today?
Don Gossen: Absolutely. One of our main focus areas is decentralized energy networks, particularly in Europe. Here's how it works. You have a local energy community—maybe it's a neighborhood or a small town. Some homes have solar panels, some have battery storage, some just consume energy. Traditional grid management is very centralized and inefficient. With our system, AI agents manage energy at the local level.
Each home or building has an AI agent that monitors consumption, production, and storage. These agents negotiate with each other to optimize energy use across the community. If your solar panels are producing excess energy and your neighbor needs power, the agents can negotiate a transaction—your agent sells energy to their agent, blockchain handles the payment and settlement, and the transaction is recorded immutably. This all happens in real-time, automatically, without human intervention.
Seth Earley: What kind of results are you seeing?
Don Gossen: In Europe, we're seeing local energy communities reduce their costs by 30-40% through AI-driven optimization and peer-to-peer trading. This isn't theoretical—it's happening now in multiple markets. And blockchain is what makes the trust and settlement possible. Without blockchain, you'd need a trusted central party to handle all the transactions and track who owes what. With blockchain, it's all automated and transparent.
Chris Featherstone: That's impressive. What are the main challenges you've faced in implementing this?
Don Gossen: There are a few big ones. First, there's the integration challenge. Decentralized blockchain systems and centralized AI systems don't naturally play well together. You have to design carefully to get the benefits of both. Second, there's regulatory complexity, especially in energy markets. Different countries, even different regions within countries, have different rules. Third, there's the education and adoption challenge. People understand AI at a high level, and they understand energy, but explaining how blockchain fits into the picture requires some work.
Seth Earley: Let's talk about data ownership. This seems like another key area where blockchain and AI intersect.
Don Gossen: Yes, absolutely. Right now, if you create data that gets used to train an AI model, you typically have no control over it and get no compensation. Big tech companies vacuum up data, use it to train their models, and monetize those models—but the people who created the data get nothing. Blockchain enables a different model. You can tokenize data assets. As the creator, you maintain ownership and control. If someone wants to use your data to train a model, they need to pay you. The blockchain enforces those access rights and handles the payment.
This is particularly important for AI training data. If we want high-quality, diverse data sets, we need to incentivize people to share their data. Blockchain provides that mechanism. And because everything is recorded on chain, you can see exactly how your data is being used and receive fair compensation.
Chris Featherstone: How does this work with privacy concerns? People don't want their personal data exposed on a public blockchain.
Don Gossen: Great question. You don't put the actual data on chain. What you put on chain is metadata—information about the data, access controls, licensing terms, usage records, and payments. The actual data can be stored off-chain in secure, encrypted storage. The blockchain just manages access rights and tracks usage. So you get the benefits of decentralization and transparent accounting without compromising privacy.
Seth Earley: What about the energy consumption of blockchain? That's been a major criticism, especially for proof-of-work systems.
Don Gossen: Yes, that's been a legitimate concern for older blockchain systems like Bitcoin. But the technology has evolved dramatically. Modern proof-of-stake systems use a tiny fraction of the energy. Ethereum, for example, reduced its energy consumption by about 99.95% when it moved to proof of stake. The blockchains we use for these applications are energy-efficient. In fact, the energy saved through optimized energy management far exceeds the energy used to run the blockchain infrastructure.
Chris Featherstone: Let's talk about the future. Where do you see this going in the next 5-10 years?
Don Gossen: I think we're moving toward an economy where AI agents are first-class economic participants. Right now, all economic activity ultimately traces back to humans. Humans own things, humans make decisions, humans conduct transactions. But as AI becomes more capable and more autonomous, we're going to see AI agents that can operate independently within certain bounds. They'll be able to own resources, make decisions, enter into contracts, and transact value—all without constant human oversight.
This requires new infrastructure. You need systems for identity, for trust, for provenance, for payments, for contracts. Blockchain provides that infrastructure. So I see a future where blockchain is the economic operating system for AI agents. It's the foundation that allows AI to participate in markets, to collaborate, to create value, and to be held accountable.
Seth Earley: That's a fascinating vision. What should organizations be thinking about now to prepare for that future?
Don Gossen: I think the key is to start experimenting. Don't wait for everything to be perfect. Pick a specific use case where the combination of AI and blockchain could create value—maybe it's supply chain transparency, maybe it's data monetization, maybe it's autonomous systems management. Start small, learn, iterate. The technology is mature enough now to deliver real value, but it's still early enough that there's a lot of room for innovation.
Also, think about how your data and AI assets are managed. Do you have good provenance? Can you audit your AI systems? Are you creating the infrastructure for more autonomous operations? These questions are going to become increasingly important, and blockchain offers answers.
Chris Featherstone: Any final thoughts for our listeners?
Don Gossen: I'd say keep an open mind about blockchain. There's been a lot of hype and a lot of skepticism, and some of that is warranted. But underneath the noise, there's really powerful technology that solves real problems, especially when it comes to creating trustworthy, auditable, autonomous systems. As AI becomes more central to how we work and live, having that infrastructure in place is going to be critical. So it's worth taking the time to understand how these pieces fit together.
Seth Earley: Don, thank you so much for your time today. This has been a fascinating conversation. For our listeners, if you want to learn more about Nevermined and Don's work, visit nevermined.io or connect with Don on LinkedIn. Thanks for tuning in to the Earley AI Podcast, and we'll see you next time.
Don Gossen: Thanks for having me. It's been great.