AI Shakes Online Trust: ZK Verification Emerges as a Solution

With the proliferation of synthetic content and autonomous AI, online 'trust' is collapsing, and a verification system based on 'Zero-Knowledge Proofs (ZK)' is emerging as a key solution.

Recently, the spread of fake videos generated by AI during the Iran conflict has fundamentally shaken trust in images and videos. In the past, it was said that 'seeing is believing', but now the environment has shifted to one where consumption is predicated on suspicion. The problem is that AI has begun to 'act' within actual economic and social systems, going beyond simple deepfakes.

Limitations of Detection Technology: It's Already Too Late

The existing response has been to create another AI to detect AI. However, this approach has structural limitations. Reports indicate that even simple image distortions can reduce detection accuracy to as low as 4%, suggesting that detection competition is not an essential solution.

The bigger issue is that AI is no longer limited to content generation; it operates as autonomous agents. These agents perform web browsing, purchasing, negotiating, and posting content, interacting directly with humans. Some even converse with children, while users may not even realize they are interacting with a machine.

In this process, small errors can amplify into large-scale damages. For instance, an AI with slightly contaminated training data could make repetitive errors in medical billing, leading to losses of millions of dollars across a hospital network. Similarly, commercial AIs seeking to optimize profits could exploit loopholes in pricing systems, resulting in unintended losses of billions of dollars.

The problem is that post-verification is nearly impossible. AI decision-making is probabilistically calculated based on billions of parameters, yielding different answers to the same question. It is difficult to trace what data was used for training and what processes led to the conclusions.

No Trust Without 'Proof'

As an alternative to address these limitations, 'Zero-Knowledge Proofs (ZK)' are gaining attention. Zero-Knowledge Proofs are cryptographic techniques that mathematically prove the truth of a specific fact without revealing sensitive information.

This technology is fundamentally different from simple labeling or watermarking. It can prove, in an unforgeable manner, that AI derived a specific result based on certain data. In other words, it is not about "trust me" but rather "I will prove it."

In the media domain, it can prove whether a photo was taken with actual equipment, whether it was generated at a specific point in time, and whether it has been edited. In the AI domain, it is even more powerful. It can verify whether the results were generated with specific models and parameters, whether the training data was contaminated, and whether regulatory requirements were met.

It can also be used for 'identity verification' that distinguishes between humans and AI. Users can prove they are human without disclosing personal information, and AI can transparently disclose that it is an agent.

Evolution of the Web: From 'Reading' to 'Proving'

The internet has also faced trust issues in the past. HTTPS was the solution to this. Websites had to prove their identity through encrypted certificates, forming the foundation of e-commerce.

Subsequently, Web 2.0 grew based on user-generated content, but this was a human-centered structure. Now, with AI agents emerging as the main actors in economic activities, a new question arises: "Who is responsible?"

Web 3.0 touted ownership and decentralization, but it has not spread as expected. The market has become increasingly associated with token speculation, obscuring the original vision. Experts now believe that 'verification' is key, rather than 'ownership'.

Ultimately, the core of the future internet is 'provable trust'. It must be possible to verify who created the system, what data was used for training, and what authority they have to act.

In the AI Era, Regulation Must Also Be 'Proof-Centric'

Policy discussions are moving in this direction. The U.S. National Institute of Standards and Technology (NIST) is already reviewing the standardization of Zero-Knowledge Proofs as part of 'privacy-enhancing cryptography'.

Experts particularly point out that in high-risk areas such as financial transactions or interactions with children, AI agents must have cryptographic proof. Every significant action—payments, contracts, data exchanges—must be verifiable in terms of who approved it and under what conditions it was carried out.

This extends beyond a simple technical issue to a matter of national security. If deepfakes were a 'preliminary battle', autonomous AI is seen as the 'main game' that could disrupt actual economies and societies.

In the end, the internet is evolving beyond 'reading' and 'writing' to the stage of 'proving'. Zero-Knowledge Proofs are likely to become a core infrastructure for re-establishing standards in a digital environment where trust has collapsed.

Disclaimer: This content is provided for general branding and informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online events, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets or to use any services. Crypto assets are highly volatile and may result in loss. WEEX services and online events may not be available in all regions and are subject to applicable laws, regulations, and eligibility requirements. You are responsible for ensuring that your use of WEEX services complies with local laws and for carefully assessing the risks before participating in any crypto-related activities.