Good morning Hunters! 🎯 AI is everywhere now, and so are the issues businesses face when implementing it. The most common one is the lack of transparency and understanding of why LLM-based products (like AI chatbots & assistants) behave the way they do. Why did it give that response? Why did it send the wrong link? Usually, it’s attributed to the so-called “hallucinations”. But what we’ve actually seen over the last 4 years of implementing conversational AI into businesses is that often the cause lies elsewhere: in messy, unorganized knowledge bases, which result in generating inaccurate or just plain wrong responses. To address this and give users more control over their AI, we’ve added a new, big and widely requested feature… I’m happy to say that we’re introducing: Message Sources 🚀 Message Sources reference the exact article from your Knowledge Base that was used to generate a particular response. With huge, extensive knowledge bases it’s hard to spot the exact bit of information that influenced the AI's response and to understand its impact. Refining this data is a crucial step in improving AI performance, and we hope we've made the process a bit easier for everyone. Happy to hear about your experiences and other problems with today’s AI products! I'd also like to take this opportunity to say that this is just the beginning of what we aim to accomplish. Actually, we’ve already started working on the next iteration of this feature, which will go even further by citing the exact paragraph used. It will also redirect you to the relevant section of your knowledge base with a click to avoid the hassle. Coming soon! If the above caught your attention, I’ll put below some more general information about Quickchat AI to give you more context. 👉 What is it? The character limit is merciless, but the PH description captures it quite well. We’re building a platform that helps companies implement conversational AI into their workflows. 👉 Who is it for? We’ve seen the most success with medium-sized companies that don’t have the resources to build their own AI systems from scratch, but have enough to treat it seriously and work with us to perfect their AI Assistants. As with everything, the more heart you put into it, the better results you get. We're focused 100% on serious business applications. 👉 What can I use it for? The most common use case is automating answering customers’ questions via a deployed web widget, but we’re also working with e-commerce and retail brands to boost product discovery and website conversion rates. We also did some cool and unusual projects like powering humanoid robots with our conversational AI engine and a few other stories I’ll save for another day. 👉 What are our key features? We put a lot of effort into improving everything that influences the quality of our AI’s responses, so I’d say that the quality is our main feature. Haha, but don’t worry, I won’t leave you with this vagueness. Here are some of our differentiating specific features: 🤝 Automated Human Handoff: Allows your AI Assistant to automatically detect and transfer conversations to a human agent when necessary. 🌐 Custom Translations: Set specific translations and exclude certain words from being translated such as brand names. 💰 Smart Lead Generation: Collect and manage data shared with the AI Assistant by users, such as contact details. All naturally during a conversation. ✨ Message Sources: Of course, I couldn’t skip that one. Message Sources give you the information on which source (article, webpage etc.) was used to generate a particular AI response. See the full feature list on our pricing page (https://www.quickchat.ai/pricing) below the available subscription plans. And since I’ve just mentioned the pricing, I’m happy to give you guys a 20% discount, if you decide to test one of our paid plans. 1️⃣ Oh, and remember that all plans come with UNLIMITED ♾️ messages (yes, even the $30/mo. plan). 2️⃣ Oh #2, before you do that, take advantage of our 7-day free trial to get a feel for our product. 3️⃣ Oh #3, and I encourage you to use our Onboarding Assistant on our homepage. You can create your account and the first version of your own Assistant just by having a conversation with AI about your needs and requirements – it’s a really cool and fast way to get started! So, I suppose I should wrap it up with a CTA now. Here it comes: Start building your own, custom AI Assistant for free at quickchat.ai! 🤖 Thanks!

Bravo on launching a tool that improves AI transparency! Congrats on your launch.

congratulations piotr and dominik on launching quickchat ai. the message sources feature sounds like a practical solution for enhancing ai transparency. can you share how this feature has impacted the early adopters of your platform in terms of operational efficiency or customer satisfaction?

great idea!