Vitalik Buterin Says X’s Grok Is Good for Truth

“When the most powerful AI systems are governed by one company, algorithmic bias becomes institutionalized knowledge,” Okamoto said. “At that point, bias stops being a bug. It becomes the logic of the system.”

Grok, developed by Musk’s xAI, is one of the world’s most widely used AI chatbots. With more than 1 billion people interacting with AI tools, a flawed or biased response has the potential to mislead at scale.

Buterin Calls for Algorithmic Transparency

Buterin has previously pushed for more transparent AI systems. Last week, he argued that developers should use zero-knowledge proofs (ZK-proofs) to verify algorithmic decisions without revealing private data.

These cryptographic tools, he said, would help users confirm that systems act fairly. “ZK-prove every decision made by the algorithm,” Buterin urged. He also recommended on-chain timestamps and delayed code releases as ways to build public trust.

Together, these tools would make it easier to verify whether platforms like X are working as intended. They could also reduce the risk of manipulation or bias, especially in AI systems that serve millions of users.

Vitalik Buterin And X

AI Platforms Face Broader Trust Issues

Grok is not the only chatbot facing scrutiny. OpenAI’s ChatGPT has been criticized for biased or factually incorrect outputs. Meanwhile, Character.ai recently came under fire for enabling disturbing conversations with minors.

Buterin believes these examples show how important it is to design systems that encourage truth over confirmation bias. While Grok still has issues, he sees it as a step forward in making X more aligned with reality—even if it occasionally embarrasses the user who summoned it.