What's up, everybody? Brian Tong here. Your guide to all things Salesforce. And this is Salesforce Explained. It's a series where we talk Salesforce and technology without the jargon or buzzwords. Now make sure that you subscribe and turn on your notifications so you never, ever miss an episode. Alright today we're talking about artificial intelligence. Otherwise known as AI. You know the buzzword that you've been hearing everywhere. So I'm going to walk you through AI and Salesforce's latest innovations with Einstein. And by the end of this video, you'll be a real Einstein yourself without all the crazy hair. I'll keep the crazy hair. Alright so let's start off with a no jargon definition of AI. AI is the science of training machines to help perform tasks that humans would normally do. In a nutshell, AI can automatically understand, analyze, and learn from data through algorithms that were specifically designed by humans. These machines are trained to recognize behavior patterns and adapt to follow those behaviors. Now, there is a lot more that goes into this, but we're just covering the basics here. So let me give you a few examples. Now, whenever you use your favorite GPS app, let's say you talk to a personal assistant, like Alexa or Siri, or you get a Spotify playlist recommendation, those are all ways that you already use AI every day. And when companies use AI tools they can automate repetitive or time consuming work so teams can focus on higher level strategic work that adds more value. They don't have to hunt down and wait for information. And they can predict things, like whether a lead is likely to convert or whether a customer is likely to stop using your product. Because your team's become more productive, businesses can reduce costs, increase revenue, and grow faster. Now, Salesforce has been actively researching and developing our own AI for a long time and we haven't just been a player in the AI field, we've been changing the whole game of AI for businesses around the world. So let me introduce you to Einstein. Hey, buddy, what's up? See, it's Einstein! Einstein is the first trusted AI for CRM, and it's built directly into Salesforce products. Einstein’s very first wave of capabilities were what's called “predictive AI.” Now, predictive AI uses existing data to provide suggestions, insights, and intelligent predictions. So this means that Einstein could figure out what worked in the past to help make decisions that'll give you whatever outcome you want in the future. For example, if you're trying to close a deal, Einstein can tell you the next best action to take. Or if you're in commerce, Einstein can look at a customer's past purchases and tell you what they're most likely to buy next. Now, when Salesforce launched Einstein in 2016, using what's called machine learning, we basically gave companies a data scientist for every team member. Like I said, game changer, right? Now, our vision was to create an AI that didn't require a ton of coding and to make it accessible to anyone and everyone using Salesforce. To do that, we brought AI, along with customer data, into our CRM, all while making sure that data stays secure and protected. Today, Einstein runs 215 billion predictions a day, and there are more than 60 specific ways that you can use it in Salesforce products. Yeah, that means Einstein works with all of Salesforce CRM products, so that's sales service, marketing, Slack, and the list goes on and on. And that makes it a key part of the Salesforce Customer 360. And if you need a little crash course on Customer 360, because I predicted that you would, you can find the link to that video in the description below. So we've covered what AI and Salesforce Einstein are. Now, let's just kick it up a notch with generative AI. Now, like we talked about predictive AI can use data that's already out there to learn patterns and make predictions about future outcomes. Generative AI can do all that and create new original content, like text, images, code, voices, and even videos. Now, fair warning - we're about to get a little computer sciency here for a minute, so stick with me, okay? Now, generative AI is a family of learning algorithms. It can also access way more data, way faster than before, and it's built with large language models, also known as LLMs or foundation models. These are very large neural networks that use machine learning techniques. They're different from traditional neural networks, which are used to solve AI problems, such as making predictions or recognizing patterns. Now, large language models include billions of parameters representing an order of magnitude increase over traditional ones. LLMs have made it possible for AI to take in a massive amount of information to train itself and understand the relationships between data so it can generate content. Just like the human brain, generative AI mimics the way that we think and learn based on examples and feedback. Now, you've already met Einstein, the first trusted AI for CRM. Well, Einstein got a lot smarter when we enhanced it by launching the world's first trusted, generative AI for CRM. Now Einstein takes productivity to a whole new level. It uses large language models, powered by your customer data, to create content. And this means that no matter what Salesforce product you're using, Einstein can now analyze all your customer data to provide answers, predictions, insights, and content in an instant. Yeah. Think of it as an intelligent assistant that's always right in your back pocket. Right back here. Ready to help out. So how does Einstein Generative AI work? Well, let's break it down into four main steps. Connect, prompt, generate, and wow. So first up is connect. Before you can create content that you can actually use, you need quality data. So we have prebuilt connectors that allow you to bring external data into Salesforce and make it available to Einstein. With Data Cloud, you can connect, cleanse, and prepare all your external data to provide to the LLM. Okay, next up is prompt. Now, after your data is connected and stored in Salesforce, Einstein can start understanding it. Now, what makes Einstein Generative AI so powerful is that it uses your customer data, whether it's inside of Salesforce or from external sources, to provide context to the large language model, so the response it generates is relevant to your business. It does this by grounding the prompt, a.k.a. a set of instructions for the LLM and all of that data that you just connected. Now, behind the scenes, Einstein uses various grounding techniques to ensure every response is trusted using real data. But wait., right? Aren't there a lot of concerns about security when it comes to generative AI? Yes, and that's why Salesforce implemented the Einstein Trust Layer. This Trust Layer is naturally embedded into the Salesforce platform, so when a request to the Trust Layer is made from any of our Salesforce apps, a number of security measures are taken to make sure our sensitive customer data is masked, or covered up, and then forgotten by the AI model after the output is generated. So the prompt and the data are never stored outside of Salesforce. Even though Einstein does the heavy lifting, there's always a human in the loop who can accept, modify, or reject the AI's output before anything is sent to their customer. Which brings us to our third step - generate. And this is the exciting part okay. As Einstein learns everything it needs to know about your company and customers, it can use that information to generate content like emails, outreach, offers, and more. So if you're a customer service rep, Einstein can provide recommendations on how to reply to a customer while you're chatting, whether it's over the phone, messaging on Facebook, or any other channel. And if you're a sales rep, Einstein can help you reach out to prospects with automated and personalized messages. If you're a programmer looking to save time, you can tell Einstein to generate Apex code for you. And if you have a commerce website, Einstein can write product descriptions that you can easily edit and publish. I mean, just think about your job right now, right? What tasks could you have Einstein take off your plate? I can think of a lot, but for me I could definitely use some help coming up with better jokes, right? Yeah. You think my jokes are funny. Because they are. But last but not least is - wow! Okay, all this data, AI, CRM, and trust stuff. This comes down to one thing and it's your customers. To build stronger customer relationships, you need to impress them with experiences they love. Now, all the previous steps we just talked about lead up to Einstein’s ability to personalize every experience and every piece of content. And this gives you the ability to wow your customers. Wow. So, okay, that was obviously a lot, right? Here we go. Let's just do a 30 second recap to make sure this AI stuff sticks. Are you ready? I guess I'm ready. Okay, start the clock. Let's go. Salesforce Einstein incorporates predictive AI and generative AI. Predictive AI uses existing data to predict future outcomes and help you make better decisions. And generative AI adds on the ability to create original content like text, images, and video. When companies combine data, AI, CRM, and trust together, they can increase revenue, strengthen customer relationships, and focus on adding real business value - all so your customers enjoy highly personalized experiences. Yes! Yes, I made it. That's why I do this. I'm so good at this. Yeah. Thanks. Am I tooting my own horn? I guess I am. So, okay, everybody, hey. That’s it for this episode of Salesforce Explained. If you want to learn more, check out the link on your screen to check out the free streaming service Salesforce+. We'll also link more resources for you in the description. Well, we'll see you next time. Bye, guys.