Aug. 23, 2025

How Human Knowledge Powers AI: Data Quality, Bias & Blockchain Payments | Rowan Stone

In this episode of An Hour of Innovation podcast, host Vit Lyoshin welcomes Rowan Stone, CEO of Sapien, to explore how human knowledge powers the future of artificial intelligence.

AI is only as good as the data it’s trained on, and while algorithms and computing power can be scaled with money, access to high-quality, diverse, and unbiased human data remains the biggest bottleneck in AI. Rowan explains how Sapien connects enterprises building specialized AI models with a global network of contributors - from doctors and engineers to musicians and everyday people - who provide the data needed to train smarter, safer, and more reliable systems.

The conversation covers critical topics shaping the future of AI, including:
* Why data quality matters more than algorithm design.
* How *bias enters AI systems and what can be done to mitigate it.
* Real-world examples, from autonomous vehicles failing in new cities to doctors using AI to improve diagnostics.
* How blockchain technology enables fair, instant, and global payments to contributors.
* The long-term role humans will play in guiding AI as it becomes more advanced and more physical, from robotics to autonomous machines.

Rowan also shares insights from his own career journey, moving from the energy industry into crypto, co-founding ventures acquired by Coinbase, and eventually leading Sapien’s mission to democratize access to AI’s growth. His perspective highlights both the opportunities and challenges of building AI systems that truly serve humanity.

Whether you’re curious about the mechanics of training AI, the risks of bias, or the evolving relationship between humans and machines, this episode offers a thought-provoking look at how data and people will define the future of artificial intelligence.

Show Notes

Rowan Stone is a seasoned entrepreneur whose career spans energy, crypto, and AI. He co-founded multiple ventures in the blockchain space, including projects acquired by Coinbase, where he spent three years leading initiatives around stablecoins, tokenization, and Layer 2 solutions. Today, as CEO of Sapien, he focuses on sourcing and structuring human data for specialized AI models used in fields like autonomous vehicles, robotics, and healthcare. His mission is to democratize access to AI’s growth by enabling people everywhere to monetize their knowledge while ensuring enterprises get the high-quality data they need.

Support This Podcast

To support our work, please check out our sponsors and get discounts: https://www.anhourofinnovation.com/sponsors/

Takeaways

AI is only as smart as the quality of data it’s fed.
The true bottleneck in AI is data, not algorithms or compute.
Everyone’s knowledge, from experts to everyday people, has value for training AI.
Global diversity in data contributors is key to reducing bias in AI.
Politically correct outputs can still be factually wrong, as seen in the Viking ship AI fail.
AI can enhance doctors’ diagnostic power while creating new income opportunities.
Self-driving cars still struggle because training is highly city-specific.
Incentive systems like staking and slashing help ensure high-quality data for AI.
Stablecoin payments enable instant, global compensation for AI contributors.
Human input will always be needed to handle real-world randomness and context.