Responsive Navbar with Toggle Menu

AI has a trust problem — Decentralized privacy-preserving tech can fix it

Opinion by: Felix Xu, co-founder of ARPA Community and Bella Protocol

AI has been a dominant narrative since 2024, however customers and firms nonetheless can’t utterly belief it. Whether or not it’s funds, private knowledge or healthcare choices, hesitation round AI’s reliability and integrity stays excessive.

This rising AI belief deficit is now some of the vital obstacles to widespread adoption. Decentralized, privacy-preserving applied sciences are shortly being acknowledged as viable options that supply verifiability, transparency and stronger knowledge safety with out compromising AI’s development.

The pervasive AI belief deficit 

AI was the second hottest class occupying crypto mindshare in 2024, with over 16% investor curiosity. Startups and multinational firms have allotted appreciable sources to AI to develop the know-how to individuals’s funds, well being, and each different facet.

For instance, the rising DeFi x AI (DeFAI) sector shipped greater than 7,000 tasks with a peak market cap of $7 billion in early 2025 earlier than the markets crashed. DeFAI has demonstrated the transformative potential of AI to make decentralized finance (DeFi) extra user-friendly with pure language instructions, execute complicated multi-step operations, and conduct complicated market analysis.

Innovation alone hasn’t, nonetheless, solved AI’s core vulnerabilities: hallucinations, manipulation and privateness considerations.

In November 2024, a consumer satisfied an AI agent on Base to ship $47,000 regardless of being programmed by no means to take action. Whereas the situation was a part of a recreation, it raised actual considerations: Can AI brokers be trusted with autonomy over monetary operations?

Audits, bug bounties and pink groups assist however don’t remove the chance of immediate injection, logic flaws or unauthorized knowledge use. In line with KPMG (2023), 61% of individuals nonetheless hesitate to belief AI, and even trade professionals share that concern. A Forrester survey cited in Harvard Enterprise Assessment discovered that 25% of analysts named belief as AI’s greatest impediment.

That skepticism stays robust. A ballot carried out at The Wall Road Journal’s CIO Community Summit discovered that 61% of America’s prime IT leaders are nonetheless experimenting with AI brokers. The remaining have been nonetheless experimenting or avoiding them altogether, citing lack of reliability, cybersecurity dangers and knowledge privateness as prime considerations.

Industries like healthcare really feel these dangers most acutely. Sharing digital well being data (EHR) with LLMs to enhance outcomes is promising, however it is usually legally and ethically dangerous with out hermetic privateness protections.