Disclosure: The views and opinions expressed here belong solely to the author and do not reflect the views and opinions of the crypto.news editorial team.
During this year’s flagship robotics conference, six of the most influential researchers in the field met to discuss a simple but important question: Will data solve robotics and automation? This gathering sparked a debate between two camps: the scale optimists, who argued that huge demonstration data sets and gigantic models would finally give robots something like common sense, and the theory’s proponents, who insisted that physics and mathematical models give meaning to data and are essential to true understanding.
Both camps are essentially right in what they emphasize, but they rarely mention the assumption that they can even trust the data they feed into these systems. As robots begin to move from the premises of carefully controlled factories into homes, hospitals, and streets, this assumption becomes dangerous. Before we argue about whether data will solve robotics, let’s address a more pressing question: Without verifiable, tamper-proof data lineage, will robotics actually destroy artificial intelligence?
When Robotics Leaves the Lab, All Assumptions Are Broken
AI continues to struggle with distinguishing fact from fiction. A recent study from Stanford University found that even 24 of the most advanced language models still cannot reliably distinguish between what is true in the world and what a human believes to be true. This example captures the core problem: current AI systems have difficulty separating actual reality from human perception of reality.
For instance, Deloitte, a well-known accounting and consulting firm, was reprimanded twice this year for citing AI-hallucinated errors in official reports. The most recent was a $1.6 million health plan for the government of Newfoundland and Labrador in Canada that contained “at least four subpoenas that do not exist or appear to not exist.” Hallucinations in large language models are not a bug; they are a systemic result of the way models are trained and evaluated.
When Hallucinations Leave the Screen and Enter the Physical World
These limitations will become far more consequential once AI is embedded into robotics. A hallucinated quote in a report may seem embarrassing, but a hallucinated input from a robot navigating a warehouse or house can be dangerous. The thing about robotics is that it doesn’t have the luxury of “close enough” answers. The real world is full of noise, irregularities, and edge cases that no curated dataset can fully capture.
The mismatch between training data and operational conditions is precisely why scaling alone will not make robots more reliable. You can throw millions more examples at a model, but if those examples are still sanitized abstractions of reality, the robot will still fail in situations that a human would consider trivial. The assumptions embedded in the data become the constraints embedded in the behavior.
Trustless AI Data Is the Foundation of Reliable Robotics
If robotics is ever to function safely outside of controlled environments, it will need more than just better models or larger data sets. It requires data that can be trusted, regardless of the systems that use it. Today’s AI fundamentally treats sensor inputs and upstream model outputs as trustworthy. But in the physical world, this assumption collapses almost immediately.
Pantera Capital’s $20 million investment in OpenMind, a project described as a “Linux on Ethereum” for robotics, reflects a growing consensus: If robots are to work collaboratively and reliably, they need blockchain-powered verification layers to coordinate and share trusted information. As OpenMind founder Jan Liphardt put it: “If AI is the brain and robotics is the body, coordination is the nervous system.”
Trustless data directly closes this gap. Instead of taking sensor readings or environmental signals at face value, robots can verify them cryptographically, redundantly, and in real-time. When every location measurement, sensor output, or calculation can be proven rather than assumed, autonomy is no longer an act of faith. It becomes an evidence-based system that can resist spoofing, manipulation, or drift.
Verification fundamentally rewires the autonomy stack. Robots can compare data, validate calculations, create evidence of completed tasks, and verify decisions when something goes wrong. You stop silently inheriting errors and start proactively rejecting incorrect input. The future of robotics will be unlocked not by scale alone, but by machines that can demonstrate where they have been, what they have perceived, what work they have done, and how their data has evolved over time.
Mark Levin is co-founder of XYO Network and operations manager of XY labs. Markus co-founded XYO Network in 2018, establishing it as the first human-powered decentralized project that directly connects real physical world data with blockchain smart contracts and other digital realities. XYO has become one of the world’s largest node networks and is experiencing record-breaking growth year after year.
For more information on the future of robotics and AI, visit https://crypto.news/robotics-break-ai-unless-we-fix-data-verification/
