Onchain Reviews - Trust Revolution or Moderation Nightmare? Trustcrypto Podcast #1
February 9, 2025
Podcast transcription:
[Intro]
🎤 Host Alan:
"Welcome to first Trustcrypto podcast. I'm Alan, and today we're diving into a question that could change how we think about online reviews. My guest today is Rachael Tavrel, Web3 enthusiast and researcher. She has been working on onchain review system and have some insights into the problems. Rachael, thanks for being here! Let me start with a question: How much do you trust online reviews today? Do you think most reviews are real?"
🎙 Guest Rachael:
"Hey Alan, thanks for having me! Honestly? I'm pretty skeptical of most online reviews. We've all seen those five-star ratings that feel too good to be true, or negative reviews that seem suspiciously coordinated. Studies have shown that a huge percentage of online reviews - whether on Amazon, Yelp, or even Google - are manipulated, fake, or incentivized. So, no, I don't fully trust them, and I think a lot of people feel the same way."
🎤 Host Alan:
"Right! And that's exactly why this discussion matters. So let's start with the basics: Why do people even rely on reviews in the first place?"
[Block 1: The Evolution of Trust & Reputation]
🎙 Guest Rachael:
"Great question! Reviews exist because, as humans, we don't want to make bad decisions. Whether it's picking a restaurant, buying a product, or hiring a freelancer, we look to others for validation. In the early days of the internet, trust was built through personal recommendations - on forums, in small online communities, or marketplaces like eBay where reputation was everything."
🎤 Host Alan:
"And then came centralized review platforms - Yelp, Google Reviews, Trustpilot. These platforms became the gatekeepers of trust. But they're starting to lose credibility. Why do you think that's happening?"
🎙 Guest Rachael:
"Two main reasons:
- First: Manipulation - Fake reviews are a billion-dollar industry. Amazon had to delete over 200 million fake reviews in 2023 alone, and yet the problem hasn't gone away. Businesses buy positive reviews and sabotage competitors with negative ones. Bots and AI-generated content make this even worse.
- Second: Censorship - Platforms have the power to remove or hide reviews. Sometimes, businesses pressure them to delete negative reviews. Other times, the platforms themselves manipulate ratings based on paid promotions."
🎤 Host Alan:
"That's huge! And now, AI is making things even more complicated. What happens when AI-generated content floods these platforms?"
🎙 Guest Rachael:
"AI is already making it hard to distinguish between real and fake reviews. With tools like ChatGPT, businesses can generate thousands of convincing, human-like reviews in seconds. Soon, fake reviews might outnumber real ones. The problem isn't just about trust - it's about information overload. How do we know what's real when AI can mimic human sentiment so well?"
[Block 2: Web3 vs. Web2 Review Systems]
🎤 Host Alan:
"This is where Web3 comes in. Blockchain offers an alternative - a way to make reviews transparent, verifiable, and tamper-proof. But let's break that down. What makes Web3 reviews different from Web2 reviews?"
🎙 Guest Rachael:
"The biggest difference is immutability. In Web2, reviews can be deleted or altered. On blockchain, once a review is written, it stays there forever. No company can remove it, and no one can rewrite history."
🎤 Host Alan:
"But that creates its own problems, right? What if someone posts a review they regret? Or what if a business improves over time, but its old bad reviews never disappear?"
🎙 Guest Rachael:
"Exactly! Immutability is a double-edged sword. It creates trust but also makes mistakes permanent."
🎤 Host Alan:
"That brings up another issue: moderation. In Web2, platforms can remove fake or harmful reviews. In Web3, everything stays onchain. How do we prevent spam and abuse?"
🎙 Guest Rachael:
"There are a few ideas:
- First: Skin in the game - Users could pay a small fee to post a review, deterring spam.
- Second: Reputation scores - If a user has a history of reliable reviews or history of transactions, their reviews carry more weight.
- In Third: Zero-knowledge proofs - These allow users to prove they meet certain credibility standards - without revealing private details. But moderation is still an open question. Even decentralized systems need some level of filtering."
[Block 3: Trust in the Age of AI & Bots]
🎤 Host Alan:
"We just talked about how AI can generate fake reviews, but can AI actually be the solution? Can it fight itself?"
🎙 Guest Rachael:
"Yes! AI can analyze writing patterns, sentiment, and other signals to identify fake reviews. But here's the problem - who controls the AI? If it's a centralized entity, it just reintroduces the same bias we see in Web2."
🎤 Host Alan:
"From what I know, Trustcrypto uses AI for automatic moderation of user reviews. Yes, it's centralized, but only to block explicitly harmful content - it doesn't alter or modify user reviews in any way. The review text is stored on the blockchain exactly as the user submitted it, signed from their laptop using their wallet's private key."
🎙 Guest Rachael:
"That's actually a smart approach! It balances moderation with decentralization. A system like that ensures user intent is preserved while filtering out extreme or harmful content. The key challenge is making sure these AI models are transparent - can users audit the AI's decision-making? Or at least see what was flagged and why?"
🎤 Host Alan:
"That's a great point - how do we make AI moderation verifiable and accountable?"
🎙 Guest Rachael:
"One idea is open-source AI models where the community can audit the filtering logic. AI should assist in moderation, not act as an invisible judge."
Block 4: The Future of Reviews & Reputation
🎤 Host Alan:
"Fast forward five years. What does the ideal review system look like?"
🎙 Guest Rachael:
"It needs three things:
- First: Decentralization - No single entity should control reviews.
- Second: Verifiable Identity - Without compromising privacy, we need ways to prove real users are behind reviews.
- In third: Reputation Portability - Your online reputation shouldn't be locked in one platform - it should move with you across different Web3 apps."
[Block 4: Impact on Brands & Everyday Users]
🎤 Host Alan:
"Now let's talk about how this affects businesses. With Web3, every review is permanent. There's no 'delete' button, no way to bury bad press. Will this put brands under more pressure to actually provide better service? Or will they just find new ways to manipulate the system?"
🎙 Guest Rachael:
"It's a game-changer. In Web2, companies can buy positive reviews or pressure platforms to remove bad feedback. But onchain are immutable - once a customer writes a review, it's there forever. This means businesses will have to think long-term. They can't just do damage control; they actually need to deliver good products and customer experiences because every bad review becomes a permanent part of their reputation. But here's the flip side - this also means businesses need a way to respond and prove they've improved. Maybe Web3 review platforms should allow companies to post follow-up responses."
🎤 Host Alan:
"That's interesting - kind of like an 'audit trail' for customer experience. Instead of deleting a bad review, businesses could show how they've improved over time. But what about regular users? Blockchain is still confusing for a lot of people. Is this all too complicated for the average non-tech-savvy person?"
🎙 Guest Rachael:
"This is one of the biggest hurdles for Web3 adoption. The truth is, most people don't care about the tech. They just want a system that works. Right now, asking someone to use a crypto wallet just to leave a review? That's a barrier. If Web3 review systems want mainstream adoption, they need smooth, intuitive interfaces where the blockchain part is hidden under the hood. Users should be able to interact just like they do on Web2 platforms - signing in with an apple or google account, writing a review - while everything happens onchain in the background. And education is key. If people understand why this system is better - why their reviews matter more in Web3 than on Yelp or Google - they'll be more willing to adapt."
🎤 Host Alan:
"So basically, make the experience Web2-level simple but with all the Web3 benefits running behind the scenes?"
🎙 Guest Rachael:
"Exactly! If users don't feel the complexity of blockchain but still get better trust, transparency, and ownership over their reviews, that's when we'll see real adoption."
🎤 Host Alan:
"And finally - your personal take. Would you trust an onchain review system more than Google Reviews?"
🎙 Guest Rachael:
If implemented well - yes. But we're not there yet. The concept is solid, but execution will be everything."
[Outro]
🎤 Host Alan:
“Alright, so we've covered everything - from the history of trust online, to how Web3 is trying to fix it, to the challenges of moderation, AI, and brand accountability. Now, to our listeners - would YOU trust an onchain review system? Or do you think decentralization introduces more problems than it solves? Let us know in the comments or email us to contact@trustcrypto.wtf
Huge thanks to Rachael for joining us. Thanks for tuning in to first Trustcrypto podcast. Subscribe and stay tuned for our next deep dive into Web3. See you next time!"