Decentralizing the Data Economy: How Markus Levin and XYO Are Closing AI’s Blind Spot

Co-founder Industry: DePIN, Blockchain, DecentralizedData Protocols

Markus Levin is the co-founder of XYO, a distributed data protocol transforming how location-based data is gathered, verified, and monetized. XYO’s mission is driven by a bold goal: to enable communities, especially in the Global South, to be active participants in the data economy. Users all around Africa are now generating actual value by providing location data from their smartphones via XYO’s decentralized physical infrastructure network (DePIN).

Data prejudice has turned into a pressing issue in an AI-driven society where algorithms influence choices in finance, healthcare, agriculture, and government, since it has profound human implications. Markus and the XYO team think inclusivity, a worldwide, distributed approach to data reflecting the lived experiences of billions, not only the digital elite, is the answer.

We spoke with Markus to explore how decentralised networks like XYO could help bridge the data divide and build a more ethical, accurate, and inclusive future for AI.

Position: Co-founder

Industry: Decentralised Data Protocols, Blockchain, DePIN

Location: Global, with strategic focus on Africa

Website: xyo.network

Combating Data Bias Using Decentralization

Q: Markus, could you explain how the concept for XYO first came to be and what motivated you to address the global data imbalance?

“The idea for XYO originated from a very simple but significant realization. Location data is one of the most valuable assets in the digital age; yet the systems collecting and monetizing it are often inefficient and unfair. We saw how centralized entities were controlling vast amounts of real-world data, using it for commercial gain, while the individuals and devices generating that data had no ownership, no control, and received no value in return.

We believed that a decentralized system could completely realign the incentives. With cryptographic technology, any data producer, like people, devices, or even AI systems, could prove the origin and authenticity of their data.  

For instance, AI systems are only as good as the data they’re trained on. So if most of this data comes from digital-first countries that dominate the northern hemisphere, this creates a global bias. When AI systems are trained on narrow, incomplete datasets, the results can reinforce systemic inequalities, produce flawed insights, and exclude billions from the benefits of emerging technologies. As IBM highlights, data bias isn’t just a technical issue – it’s a human one, with real-world consequences in healthcare, finance, agriculture, and beyond.

We are tackling this problem head-on through our decentralised physical infrastructure network model, we empower individuals across Africa to contribute location-based data via mobile, and earn money in return.”

💡 Tip: Founding a company often starts with a simple but powerful insight. Look for inefficiencies where value is unequally distributed—that’s a startup opportunity.

📘 Lesson: Data centralization creates blind spots in AI. Democratizing data inputs leads to more inclusive tech and fewer systemic biases.

Q: Can you describe how your decentralised physical infrastructure network (DePIN) model operates in practice, particularly for users providing data in Africa?

“As you mentioned, DePIN stands for Decentralized Physical Infrastructure Networks. This could refer to wireless networks, energy grids, or in our case, data-collection systems empowering decentralized geospatial location verification. While the term itself is relatively new, the concept behind it is not. In fact, XYO was the one of the original crypto DePIN projects, launched back in 2018, years before the category even had a name. We were already building decentralized networks of data-generating devices, rewarding participants, and validating data on-chain long before the DePIN movement became a re\cognized trend.

Now, how does our DePIN model operate in practice, particularly in Africa? To get started, users across Africa simply download the COIN app from the App Store or Google Play, create an account, and enable location services. As they go about their daily routines, commuting, walking, or traveling, the app quietly collects verified location data. By tapping a button in the app, they “geomine” tokens based on their movement. It’s a straightforward way for individuals to turn everyday movement into small but meaningful income, especially in areas where access to global digital economies is often limited.

XYO’s COIN app rewards for creating and sharing data, without needing to understand anything about crypto wallets, staking, or blockchain architecture. Over 80% of COIN users did not participate in crypto before using COIN. And over 95% of redeems are for XYO’s token $XYO. Most of them just use a Coinbase email for their redemption- We abstracted the complexities of crypto wallets and blockchain transactions because users care about results, not rails.”

💡 Tip: User onboarding should be effortless. By abstracting away crypto complexity, XYO expands its reach to non-technical audiences—a key growth strategy.

📘 Lesson: Real-world adoption of crypto tech depends on UX. Simplify interfaces and let users engage without needing to understand the tech stack.

🔥 Insight: Africa isn’t just a beneficiary—it’s a builder in this model. The DePIN movement empowers local economies through participation.

Q: In such a distributed system, how do you ensure data accuracy and integrity? Practical Effect and Applications

“Ensuring data accuracy and integrity in our decentralised model has always been foundational to our design. We approach this through a combination of cryptographic proofs, distributed validation, and incentive mechanisms.

At the core is our Proof of Origin protocol, which allows data to be cryptographically audited and traced back to its source.This is supported by our bound witness process, where multiple independent nodes validate and co-sign data events to create a consensus-driven record of trust.

For example, in the case where you need to confirm the number of attendees of an event or want to give attendees exclusive NFTs, being a distributed system is a huge strength. In XYO, you can have multiple sensors report it. Attendees report their attendance with an XYO-enabled mobile app, and they can create bound witnesses with each other, confirming their neighbors’ attendance. This makes it more difficult to spoof and easier to ensure data accuracy.

This is how our Bound Witness works: multiple devices independently observe and confirm the same event. Combined with Proof of Origin, which seals this record with a digital fingerprint that can’t be faked, the result is a secure, verifiable trail of what happened, when, and where.” 

🔐 Takeaway: Decentralization doesn’t mean chaos. Protocols like Bound Witness and Proof of Origin allow for verifiable, tamper-proof data.

💬 Tip: Design systems that assume low trust between participants. Use cryptography and multi-party validation to build credibility.

👀 Real-World Example: Location-based NFTs and verified attendance are just the beginning. Think beyond finance for DePIN applications.

Q: How is XYO assisting in including marginalized communities in the data discussion? What kind of influence have you observed thus far all over Africa?

“To date, there have been over 400,000 XYO data nodes across Africa, and more than 10 million nodes worldwide. The XYO community globally has earned more than $10 million USD for their contributions. Our users can make a good amount of money from participating, and it’s all in their pocket. This is even more true for marginalized communities, or countries where the dollar can go much farther — suddenly $5 or $10 USD a day could be actual supplemental income for a user.

Africa is at the heart of our expansion strategy, and we see huge potential for the entire decentralised physical infrastructure (DePIN) ecosystem.

💥 Key Insight: Even modest rewards become meaningful income when scaled across emerging economies. $5/day is more powerful than you think.

🌍 Lesson: Inclusion isn’t charity—it’s strategy. Empowering underserved regions creates loyal, high-impact user bases.

📈 Tip: If you’re building for scale, prioritize areas where marginal returns for users are highest. Africa is leading the next wave of digital adoption.

Q: In agriculture, urban planning, or public health, how might location data collected by XYO shape more intelligent AI choices?

“AI models are only as good as the data they’re trained on, and right now, that data is overwhelmingly Western-centric. This creates a serious imbalance.
Geographies like Africa are especially under-represented. The continent is home to 1.54 billion people, nearly 19% of the global population, yet it contributes to only a tiny fraction of the data used to train AI systems and shape digital infrastructure. Sub-Saharan Africa is entering its digital decade, with over half the population set to gain internet access, witnessing a 40% increase in unique mobile users from 2017 to 600 million. India and Sub-Saharan Africa will account for around half of new mobile subscribers globally over the 2022–2030 period according to GSMA.

Unlocking geospatial data across the world will help to prevent this imbalance from happening. XYO incentivises a large global user base to crowd-source this data. To date, users in Africa have set up over 400,000 XYO mobile data nodes on the continent. These nodes have all collected data that can then be fed into larger data sets like the ones LLMs can rely on to help shape more inclusive models.

Places where this could shine through AI are like these: AI could detect unusual gaps in farming activity across regions and predict potential drought conditions based on human movement patterns. For urban planning, AI models using XYO’s location data could identify foot traffic bottlenecks in growing cities to guide pedestrian infrastructure placement. Looking at public health, you can use irregularities to detect problems. If fewer people pass through an area normally known for a clean water source, AI can flag potential contamination or infrastructure failure.”

🤖 Insight: You can’t solve global problems with local data. Better AI starts with better global representation.

📘 Lesson: Geospatial data is a goldmine. Used correctly, it can fuel solutions in sustainability, logistics, crisis response, and more.

🔍 Tip: When building AI models, ask: who is missing from this dataset? That’s where the blind spots—and opportunities—are.

Q: How do you reconcile the idea of data dignity, which includes the quality, ownership, and respectful treatment of data, with user empowerment and incentivisation models, such as earning income from sharing personal data?

“When we talk about data dignity, it’s important to make sure that users are both respected and empowered to make their own choices. At XYO, we provide users with the flexibility to choose what data they’re willing to share, and in turn, we reward them for it. Whether they’re earning through completing a task or contributing to a survey, they have full control over what they want to engage with. If they don’t want to share a particular piece of data, or if they prefer not to participate at all, that’s entirely up to them.

The key is transparency and respect for user autonomy. We incentivize users for what they’re comfortable with, and we give them clear options to decide how they interact with our ecosystem. This doesn’t just help us collect more data—it helps us collect higher-quality, more reliable data. The more people involved, the more robust and trustworthy the data becomes. One person reporting data can be unreliable, but 100, 1000, or 100,000? That data becomes far more trustworthy. It’s about building a system that incentivizes people while putting control back in their hands. We believe that giving users this control leads to better, more accurate data—and that’s what really powers the next generation of decentralized networks.”

🧭 Lesson: Transparency + user choice = trust. That’s the formula for ethical data collection.

💡 Tip: Let users “opt-in” rather than “opt-out.” Empowerment grows when people feel agency over their data.

📊 Insight: Crowdsourced data quality improves with volume—but only if incentives align with honest reporting. Design for both.

Q: You are creating something ambitious in a highly competitive, developing sector—what personal experiences or obstacles have most influenced you along the way?

“I’m glad you asked, because the XYO team recently had an incident that brought this very topic to our attention and gave us a clear focal point for how XYO’s data can improve AI.

As part of our initiative to be accessible globally, XYO translates our site and a lot of our written media, like our blog articles, into a wide variety of languages. One of the languages we’ve chosen is Amharic, as we’ve identified Ethiopia as an important and growing global market.

Like most companies in 2025, we use machine translation for most of this work. Our team is careful to double check the translations to make sure they’re the best they can be. We discovered that some languages translate more quickly and accurately than others, and we were having a consistent issue with Amharic. In spite of being spoken by almost 60 million people, and being the official language of a major government, it quickly became clear that LLMs had not received adequate training on this language.

This is a clear illustration of the bias of big data that only decentralization can resolve. Big, centralized companies are always going to favor the data generated by the people they see as important. Decentralized systems not only show where those weaknesses are by giving power and a voice to those underserved people, they’re also capable of resolving them when used effectively. That’s what XYO aims to do.”

🚨 Real-World Wakeup: Machine learning gaps in underrepresented languages are symptoms of a larger data inequality crisis.

🧠 Tip: If AI can’t translate a major language like Amharic well, what else is it misunderstanding? Diversity in data is not optional—it’s essential.

🌐 Lesson: Translation isn’t just localization—it’s validation. It proves whether your model reflects the world as it really is.

Q: What most thrills you about the future of the DePIN ecosystem—and what part will XYO play in it?

“What excites me most about the future of DePIN is its potential to reshape how we build infrastructure, not just in a technical sense, but in how we organize trust, ownership, and opportunity on a global scale. We’re moving toward a world where people themselves become the infrastructure. Where anyone, anywhere, can contribute real-world data and earn value for doing so. That’s revolutionary, especially for places long left out of centralized systems.

But this opportunity comes with a paradox: the more decentralized we become, the harder it is to define responsibility. It’s not enough to say “we removed the middleman.” If DePIN is going to scale responsibly, we need systems that don’t just incentivize contribution, but verify, correct, and adapt in real time. That’s where XYO comes in.

We’ve been solving this problem since 2018. While many DePIN projects are just now grappling with data integrity, we’ve built from day one with Proof of Origin, Bound Witness, and a network of millions of mobile nodes designed to validate and secure real-world location data. We’re not just incentivizing participation,we’re verifying it.

The future of DePIN can’t rely on hype alone, it has to earn its place in the real world. That means resilient networks, redundant systems, and provable truth at the data layer. I believe XYO is positioned to be the backbone for that kind of infrastructure, especially as smart cities, autonomous vehicles, and AI systems demand higher standards for real-world accuracy.

We’re already seeing the groundwork laid in Africa, with nearly half a million nodes contributing verified data. That’s just the beginning. The role XYO will play is not just powering the DePIN ecosystem, it’s ensuring that it works.”

🚀 Vision: DePIN isn’t just a crypto trend—it’s a fundamental rethinking of how societies distribute power and value.

🔧 Lesson: Incentivizing data without validating it is a recipe for noise. Verification infrastructure is the secret weapon.

💡 Tip for Founders: Don’t chase hype—solve verification. The projects that scale are the ones solving for truth, not trends.