Hi, today we would like to introduce you to a new product that 360factor is testing called Enigma. In this presentation, we will go over exactly what problem we are trying to solve and how we are positioning this new product to address those problems. Before we get started, we would just like to highlight a shift in how the knowledge retrieval is working out currently. When Google came onto the scene about 25 years ago, they essentially redefined how users find information, which is listed as a three-step process. The user will search for whatever they are looking for as a step one, and when they type in the search question, the search result will be displayed as a step two. What the user has to do here is probably skip over all the sponsor searches and find which links the user decides they should click on to further their search. Step three is when the user actually clicks into the link, and then they will read the page and decide if the page contains the information they are looking for. So all in all, it is a three-step process. You do the search, you on your own decide which link to click on within the search return, and then when you visit the page is when you read the content in the page and try to understand what is on the page and see if the page answers your question. As everyone has already seen, ChatGTBT came onto the scene in 2023 and essentially introduced a new approach in knowledge retrieval. So instead of the three-step process I listed earlier, it became a two-step process where the user will ask the question and ChatGTBT will just give you the answer. So notice that you actually took away the two-step of going through the search result and reading the page, maybe pages, and replaced it with a single step where it gives you the answer directly. So you cannot argue that this process is far superior because it provides the answer that you are looking for directly without you having to sit through the search results and read the actual content on each page. Now, with this new approach being available to users, we can see how ChatGTBT will definitely be taking market shares from Google. So, are we just trading ChatGTBT for Google as we witness the shift in knowledge retrieval? I would argue that there are actually four competitive advantages that allow us, as in 360Factors and our clients, to step in and be able to take some of the market shares. So how do we do that? There's a few uniqueness to it. The first is a unique timing. The technology that is used by ChatGTBT actually hasn't existed for a while, but is truly being proven by ChatGTBT. So the technology is now readily available, and there are multiple models out there that allow us to leverage. In addition, because there are so many companies trying to provide this service, the cost of implementing the technology is actually going down quite a bit. So, timing-wise, we are actually in a very good position where both technology become readily available and the cost to implement this technology is going down. In addition, there is a unique opportunity for us to actually implement Enigma. Our clients have always been the gatekeeper of the most authoritative, embedded knowledge within their subject matter. So in a way, you can see on the left-hand side, we believe that our clients have really the information that users are looking for. I would also argue that our clients have the knowledge that ChatGTBT does not even have, because ChatGTBT has only reviewed the content that's publicly accessible. So that's on one side of the coin. Another side of the coin is there is a unique opportunity for 360Factor, because a lot of our clients have already uploaded their content or their knowledge to our OASIS LMS. So many of our clients have already uploaded their annual recordings, their assessment questions, their journals, their webinar recordings. So in a way, not only do you have the content that is most authoritative, if you happen to be an OASIS client, all this authoritative content actually is already stored in OASIS. So that's a very unique opportunity for us. Another uniqueness to the situation is a problem that ChatGTBT has that it cannot solve, which is a hallucination problem. So in case you're not familiar with the hallucination problem, ChatGTBT is designed to always give an answer to a question, even when it doesn't know the answer. So what that means is, if you ask a question and ChatGTBT is not sure what the answer is, it will still tell you an answer with 100% confidence, even though that is the wrong answer. So this hallucination problem, OASIS will be able to try to solve, whereas ChatGTBT is not even going to try to solve. This hallucination problem actually is happening less and less, but depending on the subject matter, one mistake out of 100 questions might not be a big deal, or it might be a big deal, depending on what the question is. So for example, medical, legal, or mathematical questions, where the answer needs to be precise, and the wrong answer might create a catastrophic outcome, you probably want to make sure there is no hallucination. Another problem that ChatGTBT has is that when it gives answers, it does not provide citations or references. So if your user looks at the answer and wants to explore additional content or resources, they are not going to be able to. So these are, I would say, some of the unique problems that ChatGTBT will not try to address, whereas 360 factor will try to address with Enigma. So what is Enigma? Enigma is going to be a custom chatbot that will pull against your data, and it will use the same technology as ChatGTBT to answer any user's questions. It will make sure that all the responses have links back to your website, whether the website is OASIS, or whether the website happens to be your guideline site, whether it might be your existing website, your main site with some of your content. It doesn't matter where the content is at, it will make sure the response has a link back to those original content, so the user will be able to explore further. And last but not least, Enigma will have analytics so that we can use users' data to fine-tune the content, and most importantly, provide additional education, because users' questions will reveal what your users are interested in, and the responses, or lack of, from Enigma will help you plan future education offerings. So what exactly is the process? So it's actually fairly straightforward. The step one is where we're going to collect the data. We want to highlight the data collection to stress the fact that Enigma is positioned as a completely separate product from OASIS, so that if you have content that lives outside of OASIS, you will have to basically import them into Enigma. If your content, or the content that you want to be part of Enigma, is already in OASIS, we will definitely have an API so that Enigma will be able to extract the data from OASIS. But the idea is step one is going to be the data collection process. Now, after data has been collected, the data needs to be processed. This is just a process of taking data that you have and save it and process it in a way that is usable for a chatbot. In addition, we'll make sure that all the linkage between the content that has been processed is maintained so that the processed data can be linked back to any of the original content, whether they are from OASIS or from outside of OASIS. Then there's going to be a bot setup process where we configure single sign-on if necessary, we configure whatever business rules and look and feel that's necessary so that we can create a bot before we launch the chatbot. We're going to go through each of the steps in detail just so that we are clear exactly how Enigma will actually execute this chatbot. For the data collection, it is very important that we want you to import all the beta content that you want Enigma to use to answer the question. If the content does not live in OASIS, we can work with you to import them into Enigma. If the content already lives in OASIS, we'll be able to give you a way to share and pick which product, which content in OASIS you want to sync over. We have an API between Enigma and OASIS that will make the process very transparent. The idea here is the data collection process is essentially a part of the implementation where you decide what content needs to be considered by Enigma when you answer a question. After you have decided what data is going to be used, there's going to be a process where the raw data is going to go through some secret sauce that we have and they become the Enigma data. Keep in mind that it's not a one-time thing. Over time, you want to add additional data to Enigma and you might even want to remove data from the Enigma data. Essentially, we can consider the initial launch is where you set the baseline and over time you can add or subtract data and they're all going to go through a similar process. For those teams that really care about security, when Enigma talks to the data, it will only talk to the data that's within Enigma Cloud. You can see here in this diagram, the chatbot will only talk to Enigma. Enigma will talk to the LMS data, likely OASIS, and or it will also talk to OpenAI to essentially make the responses a little bit more flowy. Keep in mind, we do not store the data anywhere outside of Enigma, but we do call OpenAI APIs with the data that is stored in Enigma. Basically, it's not stored there, but it is used as a parameter. Now, OpenAI has confirmed that they do not save any of the data we provided to OpenAI. This slide is to stress the fact that we keep all the data directly inside Enigma and we do not store your data in OpenAI, but we do take snippets of your data and call OpenAI to make the answer a little bit more structured. Step 3 for the bot setup is the onboarding process, just like the onboarding process for OASIS. Alright, so that's a lot of information that just came at you, and you might still get confused or have a hard time visualizing exactly what the Enigma product is supposed to do. No worries, we actually created a proof-of-concept chat, an Enigma chat called the Bible chat. This is an example where we have taken the verses from the Bible, and we have created an Enigma chat bot based on just those verses from the Bible. We are hoping to draw the parallel between Enigma based on the Bible content versus Enigma based on your content, so that if you start interacting with our Bible chat, you can get a look and feel and visualize exactly what Enigma is supposed to do. Keep in mind that your chat bot will have a different look and feel. It will definitely use only the content that you provided instead of using Bible verses, but the mechanics behind it will be identical. So all the chat interface, how the references are being presented, how it answers a question using fenced data where it will never go outside of your knowledge base, all the mechanics will be demonstrated using this proof-of-concept with Bible data. With that said, we have additional recordings on how the Bible chat works, and hopefully you will be able to visualize how the concept within the Bible chat can be translated into a chat bot with your data. Thank you!

Enigma

knowledge retrieval

ChatGTBT

competitive advantages

hallucination problem