Skip to content

Cart

Your cart is empty

Continue shopping
Ashu Dubey, CEO and co-founder of Gleen AI, on the Honest Ecommerce podcast
Apr 11, 20245 min read

AI Customer Support for Ecommerce: How Gleen Cuts Hallucinations

70-80% of Support Queries Can Be Automated, But Only If the AI Doesn't Hallucinate

Most ecommerce merchants who have looked at generative AI for customer support hit the same wall: the bot confidently makes things up. Ashu Dubey, CEO and co-founder of Gleen AI, built his entire company around solving that single problem first. According to Ashu, 80% of Gleen's technical work is focused on preventing hallucination, and they were the first company to publicly claim and demonstrate a zero-hallucination AI product. For serious retailers, he argues this is not a nice-to-have. An AI that gives customers wrong answers 50% of the time is worse than no AI at all.

When hallucination is properly controlled, the upside is concrete. Gleen targets automating 70% to 80% of inbound support queries, covering common questions about order status, returns, and store policies. That reduction in support volume directly lowers operating costs while freeing agents to handle the tickets that actually require human judgment.

Why You Shouldn't Build This Yourself With ChatGPT

A DIY prototype using ChatGPT is achievable over a weekend. Shipping something production-ready is a different problem entirely. Beyond hallucination, a self-built solution requires integrations with your Shopify store, your FAQ content, and your help desk platform, whether that's Zendesk, Gorgias, or another tool. Each of those integration points needs to be built, maintained, and updated whenever your store data changes. Without a dedicated development team, that work can stretch into months. The maintenance burden alone, keeping the AI current as products, policies, and pricing evolve, makes the DIY path impractical for most merchants.

How to Evaluate Any AI Solution Before You Deploy It to Customers

Ashu outlines three areas merchants consistently overlook when assessing AI tools. First, maintenance. A generative AI solution is not a set-it-and-forget-it install. It needs ongoing feedback loops so the model can correct mistakes over time. When evaluating a vendor, ask specifically how you provide corrections and how quickly they propagate. Second, security and compliance. Your customer data needs to stay protected, and if you operate in regulated industries like healthcare, HIPAA compliance becomes a hard requirement. Third, and most important, test the actual business impact. Vendors, including Gleen, will describe best-case outcomes. Your job is to verify those outcomes in your specific store environment before committing to a full rollout.

The Crawl-Walk-Run Framework for Merchants Who Are Skeptical of AI

For merchants who are not ready to let AI interact directly with customers, Ashu recommends starting with internal use only. In this model, support agents query the AI themselves, review the suggested answer, edit if needed, and then send it. Customers never interact with the AI directly. This builds team familiarity and lets you evaluate answer quality at zero risk to the customer experience.

The next step is embedding the AI inside your existing help desk interface so agents can pull AI-generated answers with a single click without switching tools. Once answer quality has been validated internally, you can begin exposing the AI to customers, but incrementally. Ashu draws a direct parallel to A/B testing: start at 1% exposure, move to 5%, then 50%, then 100%. Each threshold gives you data and confidence before the next step. The worst outcome, he notes, is doing nothing at all while competitors move forward.

What Gleen Actually Does Beyond Customer Support

Gleen is positioned as a platform rather than a single-use tool. Beyond automating support tickets, it functions as a virtual store concierge that helps shoppers find products and make purchase decisions, targeting conversion rate improvement alongside cost reduction. The platform also supports omnichannel deployment, meaning the same AI configuration can operate across your website, email, WhatsApp, and Instagram without rebuilding logic for each channel — a concern that Rep AI covers from a different angle in their episode on replicating in-store sales experiences on Shopify. For teams with multiple use cases, separate AI profiles can be created within a single account, covering support, product discovery, and even replacing on-site quiz tools.

On the compliance side, Gleen holds SOC 2 Type 2 certification, which Ashu describes as the highest compliance grade available in the US market. For merchants working with enterprise retailers or handling sensitive customer data, that certification matters in vendor evaluations.

For merchants already on Shopify, setup is designed to take under five minutes through the Shopify App Store. Searching for Gleen Accurate Answers pulls up the app, and the core integrations with platforms like Shopify, BigCommerce, Zendesk, and Gorgias are pre-built. Custom integrations outside that list are handled by the Gleen team and targeted for completion within one to two weeks.

Key Lessons From This Episode

  • Hallucination is the primary barrier to production-ready AI in ecommerce. Any solution you evaluate should have a clear, verifiable answer for how it prevents the AI from fabricating information.
  • A ChatGPT prototype can be built in a weekend, but production-grade integrations with Shopify, your help desk, and your product catalog can take months without a dedicated engineering team.
  • Always test metrics impact in your own store. Vendor claims about conversion lift and support automation rates need to be validated against your actual traffic and ticket volume.
  • AI adoption does not have to be all-or-nothing. Using AI to assist internal agents before exposing it to customers is a low-risk starting point that builds confidence and catches errors early.
  • When rolling out AI responses to customers, treat it like an A/B test. Start at 1% exposure and increase incrementally based on measured outcomes.
  • Omnichannel consistency matters. Managing separate AI configurations for your website, email, and social channels adds unnecessary complexity. Look for solutions that handle all channels from a single setup.

Hear the full conversation with Ashu Dubey, including his breakdown of Gleen's origin in Discord communities and how the product roadmap is shaped by merchant feedback, in the complete episode and transcript below.

In This Conversation We Discuss:

  • [00:40] Intro
  • [01:47] Bridging Discord and Ecommerce gaps
  • [03:08] The importance of reliable AI in business
  • [04:25] Balancing AI risks and benefits
  • [04:49] Limitations of ChatGPT integration
  • [06:09] Aligning AI abilities with customer expectations
  • [07:05] Improving your AI tool through feedbacks
  • [07:35] Safeguarding customer information
  • [08:35] Editing and customizing AI responses
  • [09:43] Incorporating AI into help desks
  • [10:26] Testing AI exposure levels on customers
  • [11:03] Seamless Setup with Gleen AI
  • [12:10] Gleen’s quick turnaround for integration requests
  • [12:27] Unified suite of AI services with Gleen AI
  • [14:17] Gleen’s special offer for Honest Ecommerce listeners
  • [15:01] The omnichannel capability of Gleen
  • [15:50] Find Ashu on LinkedIn and Twitter/X

Resources:

If you’re enjoying the show, we’d love it if you left Honest Ecommerce a review on Apple Podcasts. It makes a huge impact on the success of the podcast, and we love reading every one of your reviews!

Share

Transcript

Ashu Dubey

People should expect that AI won't be solving all your problems. 

Chase Clymer

Welcome to Honest Ecommerce, a podcast dedicated to cutting through the BS and finding actionable advice for online store owners. I'm your host, Chase Clymer. And I believe running a direct-to-consumer brand does not have to be complicated or a guessing game. 

On this podcast, we interview founders and experts who are putting in the work and creating  real results. 

I also share my own insights from running our top Shopify consultancy, Electric Eye. We cut the fluff in favor of facts to help you grow your Ecommerce business.

Let's get on with the show.

Hey everybody, welcome back to another episode of Honest Ecommerce. I'm your host, Chase Clymer. 

And today I'm welcoming the show Ashu Dubey. He's the CEO and co-founder of Gleen AI. They are a highly accurate and capable generative AI solution for Ecommerce that boosts online sales and automates customer support.

Ashu, welcome to the show.

Ashu Dubey

Thanks, Chase. Really excited to be here. Thanks for having me. 

Chase Clymer

Awesome. So you got a wonderful elevator pitch there. But just explain to me, if I use your guys' app, what do you guys help me do on my Shopify store? 

Ashu Dubey

So if you're a Shopify store owner, Gleen will help you increase the profitability of the store in two main ways. 

One, it will act as a virtual store concierge for your users, helping them find products, make a purchase decision, driving convergence on the store. And second, it will automate your support query. People asking questions like, “What is my order? What is my return? What's the return policy?” Those kinds of questions, the AI will answer. And it will automate up to 70% to 80% of your support queries, and that will reduce your support costs. 

So it helps in both ways by increasing sales and reducing costs. 

Chase Clymer

That's amazing. 

Where did you come up with the idea to work on this problem, to solve this problem? What was going on? 

Ashu Dubey

On the customer support side–our roots are on the customer support side and I've been very active in the AI discord, gaming, and web3 communities. The customer support in Discord sucks. If you have not used Discord, it will take you 20 minutes to just figure out how to go into a particular server and interact with people. That's really ridiculous in this day and age. 

So we decided to fix the customer support problem in Discord in the beginning. That's where we began. 

Then we started seeing people… There is a fair bit of overlap in Discord and Ecommerce, and some of the Ecommerce users who are buying products are also in Discord. So they started calling us, saying, “Hey, we have a store, could I use this bot in my store?” 

And we started getting more and more of these requests. And that's when we thought, “There is a bigger problem that we can go and create an impact on,” and that's how we came into Ecommerce. 

And once we came into Ecommerce, the field was wide open. The state of AI in Ecommerce is not very advanced. And we thought we had a chance and I'm glad we did. 

Chase Clymer

So in the last year, there's been a lot of buzz about generative AI, and there's been a lot of negative news about how it hallucinates or it can be biased. 

How does that play into this whole component here? Is it too early for retailers to really find the benefits of generative AI? Or have you guys found a solution to use it on? 

Ashu Dubey

So the concern around hallucination is actually legitimate. That's one of the key problems we ourselves encountered when we started going into AI and trying to fix the problem. So we oriented our whole company around an AI that doesn't hallucinate, and we fixed it.

So we were the first company to claim and actually do zero hallucination AI. And as such, we have a bit of a trust on that front that our political currency, if you will, on that front in the AI community. 

And so I think it is a deal breaker for customers. For serious businesses, you need an AI that doesn't hallucinate, it should give reliable answers, I shouldn't have to go and double check the answers that an AI is giving. That's a deal breaker. 

In addition to that, there are a bunch of other things also. And so, yes, the concerns around the negatives of AI are legitimate, but if you see the benefits that an AI can provide, if you can mitigate the risks around the hallucination and reliability, then the benefits are immense. 

The upside is unlimited while the downside is... If you can mitigate the downside. 

Chase Clymer

If I'm like a scrappy store owner out there, I have a DIY mentality. Why can't I just use something like ChatGPT in my store? 

Ashu Dubey

Yeah. And so if you start using a ChatGPT, you can get a prototype going within a weekend. But the problem is you will start running into the problems of one hallucination. So the bot will sometimes answer questions that it should not answer, or that it just makes up a fact out of thin air. 

So that's one problem. Your customers will get 50-50, sometimes correct, sometimes incorrect. 

The second problem is how do you maintain it going forward? Let's say you are a store owner, and you have to connect with Shopify. You may have some FAQs. You need to build all these integration points. 

The third is you probably are using a help desk like Zendesk or FreshDesk. In that case, you also need to connect with Zendesk so that your support agents can answer the tickets that the AI is not able to answer. 

So building all of this in a scalable way that's maintained is non-trivial. It can take you months to get there. 

Unless you have a large developer team that is dedicated to doing this, I would say do not do that. 

Chase Clymer

If I make the decision that I want to implement something like this into my store for customer support, what are some of the pitfalls I should be looking at to avoid? 

Ashu Dubey

Absolutely. I think one is around the expectations. So far, you should not expect AI to solve all your problems. 

In the beginning, when I would speak with customers, I usually lead with, “If you had a magic wand, what would you want the AI to do?” I hear a lot of wishful requirements, which is great because then I can map where the tech is and what the users want. 

But then in the very next question, I need to bring them to the ground saying, “Yes, that's great, but AI can do A.B and maybe C and D in the future, while E and F are not going to happen in the next two years.” 

And so people should expect that AI won't be solving all their problems.

People, they look at the AI solution from a point perspective; sometimes they ignore the maintenance aspect of a generative AI solution. It needs to constantly relearn. If a bot is making a mistake, you need to be able to give it feedback so that it can improve next time. 

So when you are looking at a solution, are you looking at just the point solution at the moment, or are you also looking at how you are going to use it in the future, from a maintenance aspect?

The third one is the security and regulatory compliance aspect. If you're new, it needs to be software type to your data. Your customer's data and their security and privacy is paramount. 

If you are in specific industries like healthcare, you may be looking at a HIPAA compliance solution as well. Those are the three main things. 

And finally, on the business side, I would say, always, always test the metrics impact. 

I'm a seller, and so we will tell you we can sell you the moon, but you should always test. I've also been on LinkedIn running lots of growth experiments, so what we think and what's actually just come out, they can be very different. 

So make sure you test it in your environment, in your store, to see whether it's generating the ROI that you initially expected it for. 

Chase Clymer

Alrighty. 

So if I am a retailer out there and you're talking to me and I'm just like, look, there's no way I'm ever going to have AI just automatically interact with my customers. What are you going to say to me? 

Ashu Dubey

The first thing I would say is, “I understand where you're coming from. You're not the first one. And it's totally fine to take incremental steps towards generative AI. And it's not an all or nothing proposition. So you can take small steps.” 

And what do small steps look like? One, you can use AI as an answer provider to your support agents or whoever your internal team is for internal users. 

For example, let's say a user is asking a question. You ask the same question to an AI, see the answer, and whether you like the answer or not, depending on that, you use it. Then you can edit that answer from the AI. 

So your users are not interacting with the AI, your agents are, your support reps are, they’re interacting with the AI and then editing that answer. So that will be the first step. 

The second is you basically add it as an app inside your help desk or on your email. So the live agents, they don't have to ask in a different interface. Essentially, it's part of their interface. With the click of a button, they can use the AI answer or they can decide to use their own answers or they can edit the AI answer. 

So that's literally the crawl, walk and run approach. So this is the walk, and the run approach is of course, once you get enough confidence, once you have tuned the AI to your liking and for correctness, you can expose it incrementally to answer automatic questions to the customers. 

And even then I would say take a percentage approach. So in AB testing, we always go with 1% exposure, 5% exposure, 50% exposure, and then 100% exposure to the customers. 

You could take the same approach with the automatic response to your customers. So the key is to get incremental confidence and take next steps, but do not sit and do nothing. That is the riskiest thing you will do because your competitors are going to use AI. Many of them are already using it. 

Chase Clymer

Fantastic advice. 

Alright. So what if I am listening to this episode and I'm pretty interested in checking out a Gleen or solution like it, how much lift does it require for me as a busy merchant? 

Do I have to train this data myself? How does it learn these answers? 

Ashu Dubey

Yeah. On the lift side, it depends from solution to solution and also on your needs. 

For example, with Gleen, if you are on one of these Ecommerce platforms like Shopify or BigCommerce, and you're using one of the core help desks, like for example, Zendesk or Gorgias, you really don't need to do much. You can do it in a self-survey, get going in less than five minutes. 

We have a Shopify app as well. Just search for Gleen accurate answers, and it will come up. Just go and install it. It will return everything. You don't have to do anything, and we will let you know via periodic communications when and what you need to do, if you need to do anything. 

But the goal here is to make it as easy as possible. 

That's for the sales part, but let's say you need a specific tools integration, the tool integration that we don't have. You talk to our team. We are very responsive. We make it possible in 1-2 weeks. 

Watever new integration requests we receive from our existing customers, we prioritize it all the time. 

Chase Clymer

Awesome, Ashu. 

Now, if I've not been really paying attention to the last 15 minutes of this podcast, what are the top 3 things I should remember about Gleen compared to the rest of the competition? 

Ashu Dubey

Yeah. That's a great question.

Glean does not hallucinate. The answers are extremely reliable, and that's critical for you as you evaluate different solutions. 80% of our tech is focused on preventing hallucination. We give you your own LLMs if you need one, or we default with the best that's out there. 

Glean is a unified suite of AI services. So don't just think of Gleen as either a support solution or a product sales solution, think of it as like a one-gen AI for all your Ecommerce needs that you will ever need. It's a platform where you can launch multiple AI. 

Maybe your marketing team is looking for an AI solution. Gleen can meet that. You can create different profiles within Gleen to do that. If your sales team is looking for a specific one, maybe replacing a quiz on the website, Gleen AI can do that.

And of course on the support, that's where our core strengths are. 

And finally, on the third side, Gleen is extremely compliant. We take your user data and privacy very seriously. We are SOC 2 type 2 compliant, which is the highest compliance grade you can achieve in the US. So you can trust Gleen to keep your stores data, your customers data secure and confidential.

And we worked with very large retailers already. So if you are looking for references or case studies, go to our website, check out some of the case studies. 

Chase Clymer

That's a great answer. 

And I feel you guys had a special for listeners and viewers. 

Ashu Dubey

100%. I'm glad you asked. Exclusive to your viewers and listeners, we got $250 worth of credits on Gleen’s paid plans. This means you get access to the advanced features and also dedicated support from our team. So feel free to avail that opportunity. 

Chase Clymer

We'll make sure to link to the proper program in the show notes. And again, 250 free credits for any of Glean's paid plans if you go to that link in the show notes. 

Ashu, I can't thank you enough for coming on the show today and sharing all your insights.

Before we go, is there anything I didn't ask you about that you think would resonate with our audience? 

Ashu Dubey

One thing is just... I talked about everything, but one thing I probably didn't touch on is the omnichannel aspect of Gleen. 

Usually, you don't interact just with users on one channel. You're interacting with them on websites, on email, on WhatsApp, on Instagram. Gleen is an omni channel solution so it can meet your users where they are. It's the same AI but across all the channels. So that's one thing I didn't add and it's super important. 

Chase Clymer

Oh no, that is absolutely important. Ashu, you've been a wealth of knowledge. I'm sure I'll have you back on the show talking more about generative AI in the future. We're going to have all the links in the show notes today. 

If people want to get a hold of you, Ashu, where can they find you? 

Ashu Dubey

They can find me on LinkedIn, just search for Ashu Dubey - Gleen. And also on Twitter, my handle is Ashu Dubey. Actually, my Twitter handle is dubeyism. And so just on Twitter, you can find me with that handle. 

Chase Clymer

Awesome. Thank you so much for coming on the show today. 

Ashu Dubey

Thanks Chase. Thanks so much. 

Chase Clymer

We can't thank our guests enough for coming on the show and sharing their knowledge and journey with us. We've got a lot to think about and potentially add into our own business. You can find all the links in the show notes. 

You can subscribe to the newsletter at honestecommerce.com to get each episode delivered right to your inbox. 

If you're enjoying this content, consider leaving a review on iTunes, that really helps us out. 

Lastly, if you're a store owner looking for an amazing partner to help get your Shopify store to the next level, reach out to Electric Eye at electriceye.io/connect.

Until next time!