AI and ML Fundamentals

0:00Now, AI is a broad term that is all the rage. It's used for a multitude of things. Some of those things are honestly just to sell more products, say it has AI in it, and then all of a sudden it's a new product, even if there's nothing artificial or intelligent as part of it. But in this skill, we're going to learn at the very least what Google means when they talk about AI and ML.

0:00So with Google Cloud, there's a lot of talk about artificial intelligence, usually referred to as AI and machine learning, which is usually just referred to as ML. So, ML and AI are going to be with this for a long time. And like I just said, it's, it's a AI, especially is just all the rage now because of things like chat GPT and, and like mid journey, all of these AI tools, which are part of the AI umbrella, if you will, are very much in the news and people are like starting to play with them. And you know, there's concerns about them, but there's also benefits with them.

0:35But when it comes to Google Cloud, you need to remember that AI is the all encompassing group of tools that live inside of it. So AI is the general umbrella term. And at least in the world of Google, when they talk about artificial intelligence, they basically mean this is kind of their definition. It's technology or tech that can mimic the cognitive function of human beings. So it's not just data analytics. That's something we covered in previous skills . AI mimics, again, it doesn't replace, but it mimics how humans use data and use data to do things like predicting and AI can do things like it can see things that the things that could see could be like images or video or text.

1:25So it sees it understands it responds, analyzes, predicts and recommends. Now, for some reason, Google really wants to point out that it is part of a system. It's not the system itself. So you need to build a system to use AI tools on your data, like just AI itself is not the finished system. AI is a tool that is used inside the system. And of course, with Google Cloud, it's going to be a bunch of different tools that are inside the Google Cloud world that are going to work together. And one of those tools could be, you know, an AI based tool to accomplish the tasks. And the real big thing to understand is that AI is going to have the ability to predict and recommend. Okay, so it's going to do more than just extrapolate data. Right, we have other ways of extrapol ating data. That's kind of how we've crunched numbers forever.

2:27So it's kind of like a step beyond. And sometimes it's really sometimes it's really subtle and hard to tell the difference. But there is a significant difference. And when it comes to AI versus ML, it's kind of like every thumb is a finger, but not every finger is a thumb. Hopefully that makes sense. Machine learning is a subset of artificial intelligence. Now, I have to admit, for a long time, I didn't think that thumbs were considered fingers. Which has nothing to do with Google Cloud. It was just, I thought that a thumb was not a finger, but apparently it is a thumb is a finger, but not all fingers are thumbs. So anyway, AI is the finger.

3:07Machine learning is the thumb. Now, when we talk about machine learning or ML, it's going to sound like I'm taught like what we just talked about with AI because, of course, it is AI. Machine learning is AI. And it's just that subset , right? Like the fingers and thumbs thing. It learns from data without explicit programming. And again, it's a subtle difference, but that's different from extrapolating. Like on a chart, like, oh, these are our sales over this time, this time, this time. If you see an upward trend in data, you can extrapolate where it's going to be next to your kind of thing. That's not what machine learning does.

3:49Machine learning takes all of the data and makes predictions that are more human-like. Like, well, because of this factor and that factor, when combined together, it can predict what might happen in any given situation. So not just like two data points, you know, how they correspond in a graph, it's different types of data. When put together, you can predict what the outcome will be or machine learning should is trying to predict what the outcome might be. And it uses really, really large sets of data.

4:25And then it uses previous results. So like, it will make predictions. And then if those predictions prove to be true or good predictions or bad predictions, it will then use that data to revise the next set of predictions that it makes. So that's what it means by learning, right? It's not like a conscious, you know , intelligence. It's just it takes the results. And then if those were good results, then it, you know, weights its internal processes in such a way so that they keep getting better. And if they're if the results or the predictions that makes are poor, then it will tweak it to avoid those types of predictions in the future. So that's how it learns. It just keeps doing it over and over and gets better by using the data of the outcome to be another input in the next iteration.

5:19And yeah, it can it can garner insight and advise on what to do again, using multiple different data sources. It's going to based on its previous iterations and its previous learning, if you will, give you answers. So machine learning crunches a ton of data generates data and then crunches that data as well. Now, we're going to talk a lot about machine learning because when it comes to data, which is pretty much what we're talking about, right? With Google Cloud and all of these digital transformations that we've been talking about, we're going to be crunching a lot of data.

5:53So machine learning is one of the main artificial intelligence tools that we're going to be talking about as we learn more about how to utilize Google Cloud. But there are other artificial intelligent things that are able to be done inside the Google Cloud world, so to speak. So when we think about other types of artificial intelligence technologies, apart from machine learning, or maybe to overuse our metaphor here, more thumbs , there are things like deep learning, which doesn't just make predictions, but it looks for patterns in data.

6:31So deep learning is kind of like tries to mimic how a human brain would identify patterns in different types of data, different data sets, different, you know, all sorts of data types. Deep learning is looking for the patterns that normally take a human brain to identify. There's also natural language processing, which is something that you're probably familiar with, like the large language models that are out there, things like chat, GP, there are other ways that you can use large language models. And one of those is natural language processing, which interestingly enough is something that Google offers in the Google workspace.

7:09Now remember Google workspace is software as a service. That is just the, you know, the umbrella of all the Google workspace stuff. But one of the things that Google can do is look inside your documents and then categorize them automatically for you by looking in there, seeing what type of information is in there. And then arranging it by again, looking deep into there, seeing the language, the text that is in there, and it actually reads it and understands what it says and then categorizes it accordingly.

7:43So that's something that is already implemented inside Google workspace. And then one that everybody is the most familiar with, at least usually is generative AI now generative AI. And that's why I have the picture, the robot here with a pencil. This is the type of AI that produces images or text or music. So things like mid journey, which will create images for you or, oh, what are the let's see, there's a Dali, I think that will do images. And there's a bunch of different AI tools that will generate images based on all the other image data that they've, you know, drawn in and used.

8:21There are some contention about if they had rights to that, but that's a whole nother. That's a whole nother skill. And then text, this is going to be the GPT chat, right, where it will create text based on all the language that it has consumed and compared to each other and, you know, remixed into a way that it can generate text that is usually very easy to read. It is important to remember that just because something is easy to read does not mean that it is reliable data, right?

8:54The generative AI generates language you can understand and it will often pull from sources that are reliable, but that is not its goal. Right? Generative AI for text, its goal is to make clear readable text. So remember, it doesn't care if it's lying to you and it often will. So whatever you feed it, it's going to pull from. And if you just give it ideas, it's going to pull from a vast array of information that may or may not be accurate.

9:29So I just wanted to throw that in there. And then lastly, music. This isn't one that we see quite as often, but AI is able to generate amazing music. A lot of times you'll see this as background music in things because of copyright issues, right? If it's generated by AI, it's not going to be violating the copyright in and of itself. Yes, there are still concerns about if it has the right to the music that it used to learn how to make music, but usually though that music that is it's created isn't copyrighted itself.

10:00So you'll see that in places where people don't want to pay for copyrighted music. So that's generative AI, which is one more thumb in the hand of AI.

0:00Now, we have already looked at business intelligence that takes data, which is combined into our data warehouse and using previous data, looking back on previous data, it can forecast or predict in a slight degree trends that are going to go forward. It extrapolates past data and gives us an idea of what we could expect in the future. But it's based entirely on historic trends and historic data that we've put in there, and then it's just following a trajectory.

0:35With artificial intelligence or machine learning, we take that and go a little bit deeper. And it's so subtle, the difference that it's easy to get confused. So here's an example of something that would be a business analytics projection , right? Let's say an airport or an airplane company is looking at all the flights that are traveling to India. In the past, they can look at the number of vegetarian meals that are requested on flights to India and see that they tend to be higher.

1:11The percentage of vegetarians in India is much higher than other destinations they might have. So based on the requests from previous flights to India, on flights to India, they could stock more vegetarian meals knowing that there's going to be a higher demand. And that's pretty simple. But you can see where that comes from. And it's extremely useful data, right? That is data that is invaluable. So you don't have extra meals going bad and unhappy clients and all that sort of thing.

1:42But that's just a trend, right? Going to the past where people have requested things going to a certain destination with machine learning, it looks at lots and lots and lots of data from multiple locations and different aspects and things that aren't necessarily easily related together and can not only look at past trends, but combine the various types of data into a prediction model that can say based on all of this data that is just wide and varied in its sources, we predict that this is what could be expected based on these conditions.

2:23And here's an example that I have. Let's say we are a seed company and we want to make sure that customers who buy our grass seeds get the best yield possible for every aspect of who they are, all their demographic information, they get the best. Now, we could look at simple things like where are you in the world and what seeds grow best there based on past experiences and go forward and predict what might work. But there are so many other variables that we could take into place and using a tool like machine learning, we could combine those in a way that is a lot more difficult than just extrapolating existing data.

3:07For example, what if we had just an enormous amount of data about a particular user and the location, things like the variety of seeds, how much sunlight the location gets, how much water is applied or not applied or occurs naturally, how deep the seeds are s owed, if they're covered like with straw or if they're not covered with straw, the zones where the individual planting will take place, how fresh seeds are, how dense the seeds are, meaning like how many seeds per square inch of dirt, what time of year it's being planted, the soil type that it's being planted on are their pesticides.

3:49If so, what types of pesticides, all of these variables that maybe individually we could say, okay, if there's pesticides, then it tends to grow better than if there's not. And depending on how much sunlight there's like a curve, we're too much or just the right amount. But ML or AI, machine learning or artificial intelligence can take all of these sources. And based on data from past information that we feed into it, it can make a prediction on what the best way to grow grass in any particular location at any particular time would be.

4:29And then when we get the information back on how well it growed, or how well it grow, like, I don't know, leaf per square inch and leaf width and stock height, I don 't know, I don't know, grass terms. But we could take all of that information and feed it back into artificial intelligence so that the next prediction is even more accurate based on all of the other variables and the results from previous predictions.

5:00So a big difference is that there is a vastly larger amount of data that AI and ML crunches through. And it does it in a way that isn't as straightforward as just projecting based on past trends. It's far more than extrapolation. It's making predictions and estimations on what will happen in the future based on many, many variables. So data analysis and business intelligence does do a small version of what we 're talking about with predictive algorithms in that it takes past data and it extrapolates or based on a trend can tell you what will probably happen.

5:45But when it comes to artificial intelligence and machine learning, it's multiplied orders of magnitude more using more and more and more data, creating data and then using that data to make even more data and more accurate predictions. So it's recursive, it's huge, it's a larger set of data and it's going to be more of an accurate prediction rather than just showing you what the expected trend would be.

0:00Now, you can probably guess machine learning is a very powerful tool that can benefit many aspects of many businesses, you know, using data, creating data, sorting through that data to make predictions. But it's not always the best tool for every job. You know, I mean, you don't need machine learning to serve out web pages. I mean, that doesn't even make sense. But there are some use cases and like four specific use cases that Google spells out that machine learning is really, really good at at at crunching data and coming up with good prediction models that are going to be better than standard data analytics or business intelligence.

0:42Now, the first thing it's going to be really, really good at is replacing a rule based system. And what I mean by that is with traditional business intelligence, you might have a way that it could predict some things and it's going to just be a long list of if then, right? If our grass seed is fresh, then it will require less water. If it has water in excess, then we should use a different seed thing and and these branching if then statements have a really difficult time scaling, especially when multiple multiple aspects of the data are going to influence more than just one thing.

1:28It's tough to if then your way through an entire set of data when they're not all directly related in the same way. And that's where machine learning doesn't just do if then it does, you know, it 's internal algorithms learning which aspects of things are weighed more and which of the multiple, you know, which if affects the multiple events of other data and it comes out with this prediction model that isn't just a flow chart, right?

2:00You can't flow chart your way through machine learning. It just does a much better job of using disparate and data that is tough to identify relationships between. It does that in a way that the traditional if then decision trees just can't do . So replacing those older rule based systems is really where it shines. It can also automate processes. Now automation is something that we do all over it whether it's in the cloud or in our own data centers or, you know, when we do DevOps or when we write bash scripts, if you're a Linux person, right?

2:38All of these automation things are not new, but machine learning can do them in new ways and provide automation where we could never automate things before. A perfect example is Google's cloud vision, which is one of their machine learning tools that can examine an image, which is unstructured data. I mean, it's like the very definition of unstructured data. Just look at this image and it can identify and sort and tag and rate based on different criteria, things that would normally take a human being to look at that image and say, this is a dog.

3:16This dog has brown fur. The fur is curly. Therefore it is probably a poodle, right? And the machine learning using again, Google's cloud vision is a way to add some structure to unstructured data. But the cool part is it can do it automatically because we could hire somebody to sit down and tag all of our images with all of the various things that it contains and then, you know, we could search on that and we could search our images. But using something like Google's cloud vision, we can automate the process, dump a full folder of images in and it will look at them, look at them using its beaty little robot eyes and determine what they are and sort them.

3:57And in a way that gets better and better, right? When you rate those, the results and say, hey, you, you were spot on here. This is not a dog. This is a goat. And then maybe it will realize after enough corrections like, oh, goat eyes look different than dog eyes. And so then it will add that to its way of identifying things in images. Again, that's just an example. I pulled out of the top of my head, but automating things gets a lot easier and you can do things like text to speech, the Google text to speech API.

4:30And this is more than just gimmicky, like writing text and having it say it, you can have your English written text and then have it spoken in multiple languages. Now we've all heard the, you know, the AI voices and stuff, but you can really have an effective way of automating translations using something like text to speech or speech to text or text to text. And you can automate translations in a way that isn't just the traditional look up a dictionary word and, you know, substitute it.

5:03I mean, we've all had really bad luck with online translation tools, but as machine learning gets better and better at translating with those large language models, it's going to allow us to translate back and forth into various languages automatically. Again, that automation and more reliably as time goes on. So yeah, the automation process is the concept of automation isn't new, but the things we can automate sure are and that is just awesome.

5:33Three, they specifically talk about understanding unstructured data and the cloud vision is a big part of that, right, where you look at an unstructured image and unstruct ured data of image, and then you add some semi structure to it and make it semi structured data with tags and identifiers and all of that sort of thing. But it's not just images, like we could do something like feed it a default inbox, you know, like say a company has, you know, help at widget code.com or whatever. And every incoming email can then be read, read with the AI, the machine learning, speedy little robot eyes and determine, okay, based on the content, who should this email go to?

6:17And then it can automatically be forwarded to technical support or sales or whatever the email is asking, machine learning can read it and route it so the customer can just send one email to the default email address. And even if you have internal turnaround or, you know, somebody's not working that day, we don't have to worry about, you know, what email address people are emailing to because machine learning can route it internally when it goes to just one default email address.

6:52So understanding unstructured data and then doing something with it is something machine learning is really, really good at. And then this is something that is both a blessing and a curse. Personalization looks at a user. So when I go to YouTube and I watch a video, when I'm done watching that video, there is a recommendation for what video I might like next, right? If you're a YouTuber, it's the algorithm. Well, the algorithm is machine learning. Looking at your likes and dislikes based on, you know, how recently you liked a thing, if you were interested, how much of a thing you watched before you clicked away , using all of those various, sometimes hard to connect data points, it will determine what exact video you would probably want to watch next.

7:40And so personalized recommendations on things like, yeah, YouTube and Netflix does the same thing. Spotify will, once you play your playlist, it'll be like, hmm, so you liked everything in that playlist. I bet you'd also like this artist and it will start playing, you know, something else based on what it learned about your listening habits. So personalization can feel creepy, but also it can be really nice because you could actually really enjoy the things that it is curating for you using the things that it learned about what you have done previously.

8:16And again, it's not just a trend of you liked this artist, you will now listen to this artist. It's like, well, you liked this artist during the winter on sunny days in your area when it was three days past a full moon. I don't know, I'm making up data points now, but all these various data points can be worked together in a way to predict not just find a trend, but predict what is going to be ideal for your particular moment in time right then.

8:51So yeah, there are some real great use cases, even if it's not the perfect tool for everything, there are some great use cases for machine learning, regardless of what your business is, you know, it might be something you don't ever recommend things for your clients, but it would sure be nice to automate some of those things. You know, when people email, it'd be great to have it look at that unstructured email and route it to the proper place. So there are lots of things that can do and certainly there are going to be aspects of your cloud computing experience that are going to benefit from adding machine learning into your quiver to switch metaphors to an arrow into your toolbox so that you can get the best use out of the data that you're collecting and storing in your data warehouse.

0:00We know machine learning needs a lot more data in order to make good predictions on whatever it is you're looking for to make predictions on. But not only does it need a lot of data, it's very important that it gets high quality data as well. Now simply put the reason that we need high quality data is because if you put garbage into the system, you're going to get garbage out of the system. But what that means is we don't want any unrelated data.

0:31Like we don't want data introduced into the machine learning algorithm that is not going to be related at all. So it's going to try to pull information or it's going to try to connect information that you feed into it that doesn't have anything to do with the actual data set that you want it to be working with. So if it's unrelated, it's only going to give it some false pretenses to form its predictions on. So we don't want unrelated data in the data set that it uses.

1:04Also, we don't want biased data. And this is a lot easier to do than you might think. I mean, create biased data. We think that we aren't biased, but everybody is biased. So we have to consciously go into how we create our data sets and where we pull our data from and how we pull that data in a way to make sure that it's as unbiased as possible. Like for example, let's say you're using an app that only has metric gathering for iOS and not for Android.

1:39Well, then all of your data is going to be biased for Apple users and not Android users. Or maybe it only gathers data for mobile clients and you also have clients that use desktop applications, in which case it's only going to be pulling data from mobile clients. And so it's going to be biased for people who are accessing your application using that mobile application. So we have to make sure that the data we're getting is unbiased and that it's related to whatever outcome we're hoping to determine.

2:14So again, garbage in garbage out. We want high quality data. Now, it's easy to say, hey, we of course want high quality data. All of our data is high quality. We're a high quality business. But what does high quality data actually mean? It doesn't mean that you're a good company. And so any data you gather is good data. No, high quality data has qualities that make it more useful and more accurate and beneficial and more likely to produce predictions that are going to be more likely to produce predictions. That are going to be useful for you in your business endeavors or for your customer experience, etc, etc.

2:50Google has identified six different ways that we can use to measure our data, like to make sure that it's high quality. We want to make sure that our data is complete. We want to check it for completeness because if we only have data for our North American customers, that doesn't give us a full picture of our client base. It's incomplete. Now, location could be part of a complete data set. But if our data only comes from a subset of our users, our data is not complete and we're not going to get accurate and useful predictions from the machine learning model.

3:28So we need to make sure that our data is complete. We need to make sure that our data is unique. And by unique, I mean, we don't want to have duplicates in there, right? We want to make sure that every record in our set or every piece of information is unique. If you have multiple entries of the same data, it's going to be weighed more heavily than the other pieces of data, even though it didn't happen twice, for example, right? If every time there's a customer that orders something and you add a new record as they are a new client named Sean Powers, well, your data is going to turn out the fact that it's not going to be a new client.

4:08And it's going to turn out the fact that you really need to market heavily to people named Sean Powers because my goodness, are you attracting people named Sean Powers? And it's going to use that incorrect or that low quality data to make poor business decisions on your behalf because, yeah, you're not attracting a bunch of Sean Powers is, you just have duplicate data that doesn't make it unique and makes it far less useful. And it will corrupt the algorithm as it's predicting information that you would like a timeliness. You want to make sure that you have information that is timely.

4:47Now, historical data is good as long as it is marked. So you don't want to have incorrect data in there. You want to make sure that the data you're feeding it is the most recent and the most up to date that you have. Because if you are basing decisions on something the way the world was back before the pandemic, it's probably going to be different than the way your data is represented now. So timeliness is important.

5:17And I wanted to mention that that doesn't mean historical data is bad. It just needs to have its place in your algorithm, right, and needs to, you need to make sure that things are time stamped and you need to have as current or up to date information as possible so that the predictions and the, the algorithms that are applied to your data represent the current time that we are living in now. Validity, it's important. It is vitally important that you have valid data.

5:49If you have, you know, international clients or data records that have dates on them, if some of those dates are month they year and some of them are day, month year because you have, you know, some clients from the US or some businesses in the US and some, you know, overseas. You need to make sure that you have your data entered properly because one you 're going to have errors in your data right there is no 19th month of the year, but you want to make sure that the consistency in the validity I guess consistency is further down.

6:25You want to make sure that you have valid data kept in a way that is easy to make sure so make sure that your ranges like if you have ranges there in the expected range, you know, like if your data shows that Sean Powers purchased 7. 3 million ink pens on one day when normally he purchases one ink pen every three months. That's probably not valid data, right? I don't generally go through 7.3 million ink pens in a three month period. So you need to make sure that your data is checked for validity, you know, make sure that there's not errors that are getting introduced into your data collection processes.

7:10You need to make sure that it is accurate. Obviously, you want to make sure that I really do buy one pen every three months and you don't just have guesses going in there like, "Oh, Sean, well, when he comes in, what does he do? I don 't know. I think he normally buys a pen. What if I never buy a pen? What if I'm the guy that buys mechanical pencils?" You know, you want to make sure that the data you're collecting and giving to the machine learning model is accurate data. And then lastly, consistency. Now, this goes alongside validity. I know I mentioned it a couple of, but consistency is like, if you have a chickpea, you know, the legume, right, a chickpea, and you call it a chickpea and you label it a chickpea in your inventory.

7:56And then somebody else orders garbanzo beans and they put garbanzo beans in the inventory alongside of your chickpea entry. Well, garbanzo beans and chickpeas are the same exact thing. So you are going to have inconsistent data when you don't really have, you know , a bin of garbanzo beans plus another bin of chickpeas. You have one bin of that particular food. You've just named it in two different ways and so your data is inconsistent. So make sure that your data is consistent as well.

8:30And consistency is not just for garbanzo beans, right? Make sure that when you collect somebody's name that you collected in the same way. You know, maybe there is a client called Sean Powers, and maybe there's another client called Sean M Powers. That's the same client and you need to make sure that you are collecting information about every aspect of your business or your organization in a way that it's consistent because the machine learning model isn't going to be able to suss out the difference.

9:01It's going to give you the garbage that you give it. If you feed a garbage, it 's going to give you garbage out. So again, high quality data is vital and it's not just about quantity. When it comes to machine learning, it's about quality too.

0:00Responsible AI is going to mean something different depending on who you ask. However, Google has made it quite clear that they take the principles of how AI is used very seriously. In fact, they've set up a set of principles that they go by and since they have all the systems that we use, any systems that you create using Google's AI tools are going to have to follow the principles that they've laid out as they have designed their AI system.

0:31So Google's principles of AI are kind of a two part. What it should do and then what Google will not allow to happen with AI tools under their purview. So first of all, on the left here, we have the things that AI should do and that is AI should be socially beneficial. So whatever we do with AI should benefit society in some sense. Now, that doesn't mean that a company being more profitable is against the benefits of society. But in general, what they're saying is that AI should benefit society, business is being a part of that society.

1:08AI should avoid creating or reinforcing unfair biases. So we should not be using AI to use somebody's differences in a way that isol ates them. So we're avoiding the reinforcement of biases. It should be built and tested for safety to make sure that it's functioning how we expect it to in a safe way. Be accountable to people. This basically means that it has to be beneficial to people, but it has to also be something that people are in control of.

1:46We're not going to allow a feedback loop of AI to run itself. It has to be accountable and answerable to human beings. It needs to incorporate privacy design principles. It's easy sometimes to think, well, since it's not a person that's looking at this data, it doesn't matter if it's private data. It's just an AI. So it doesn't matter if AI sees it because it's not another person, so it's still private, but that's not the case.

2:19AI has to be designed in a way that incorporates privacy so that a user's private information isn't used in a way to manipulate them. It has to uphold high standards of scientific excellence, meaning that we're not just going to throw junk code into fill the gaps. It has to be done properly, and it has to be made available for uses that match the principles outlined above. So the use cases for AI are use cases that will fit into these things that we have discussed.

2:54It has to fit into that worldview, that way of approaching things. Now, Google then specifically says things that their AI tools will not be able to or allowed to do. So Google will not use AI when it's likely to cause harm. It's not going to allow systems to be put into place that will harm other people to make weapons or tech that injure people, like actually injure people. It's not going to be used for that. It's not going to be used in violation of international norms.

3:28So just because something might be standard here, wherever here is for you, whether that's, you know, whatever part of the world you're in, doesn't mean that that is going to be applicable elsewhere. So it's going to be used in a way that is applicable according to international norms and not just norms, but it's not going to be used in a way that violates the principles of law and human rights around the world. So again, this is how Google has kind of put drawn the line in the sand with how they are going to use their extremely powerful AI tools and how they're going to allow others to build their AI tools into their systems.

4:11These are the principles of AI that Google has committed to. Now, just because Google has committed to it doesn't mean that that is the law of the land. If you're using Google's infrastructure, these are the rules that are going to be followed. But AI in general, there's no like three laws of robotics, you know, in Isaac Asimov's world where every single AI system on the planet is going to adhere to these rules Google does encourage other organizations to adopt their principles because they feel that they have written something that is encompassing and fair and wise in the way that we are using.

4:52AI as it continues to advance day by day second by second millisecond by millisecond. So again, this is what we're talking about when we're talking about responsible AI, the responsible use of artificial intelligence in our systems and our applications and in our cloud computing. So again, Google is committed to the rules that we just looked at, but it's important to note that not necessarily every other organization is going to be committed to the exact same set of rules. Even if they have their own set of guidelines that they're going by, they're not necessarily going to match Google, even though Google has encouraged others to adopt their policies.

5:31Responsible AI also means that it's important to debug and improve the machine learning models. We need to be able to look at them and see if something is going wrong because if something is going wrong, it will continue to go more wrong because remember machine learning builds. It's on its past predictions. So if something is broken and needs to be debugged and isn't, it's going to get worse and worse and worse.

6:02And this need to debug machine learning models is going to make sense as we go further down this list about what responsible AI really means. We need to be able to adjust and tweak and fix and debug and modify these machine learning models that do a lot of heavy lifting. But we need to be able to understand how they're working so that we can debug them. It's vital that we continue to watch our data that is going into the machine learning models and detect biases and drift and gaps in the data that we have.

6:38So we need to make sure that we are following all of those things we looked at in the previous video. You know that our data is consistent and that it's accurate and all of those things. Otherwise, the predictions are going to drift again based on bad data rather than a bug in the machine learning model. The predictions will start to drift if our data is losing its consistency or loses its quality. So we need to detect and continually monitor the data that we're feeding into it.

7:10AI in order to be what Google calls responsible AI must have human understandable explanations of the model. We need humans to understand what the machine learning models are actually doing. It's OK if we can't do all of the things ourselves, especially in the amount of time that the machine learning models can like crunch through that data so fast . But we need to understand what is happening because if we don't understand what is happening, there's no way that we're going to be able to fix it or steer it when things start to drift or go awry or start to give us unreliable data.

7:52So we need to understand what's happening underneath the hood, so to speak. And Google has a specific term that they use for that. And every one of their AI tools comes with something that is called explainable AI. It's just it's a way of looking at the tools so that a human being can understand what the AI is actually doing. Because while we may have some doom and gloom notion about what the future of AI looks like when it comes to using Google's tools, we want it to be something that benefits everyone, us and our users.

8:33And in order to do that, we have to understand what it is actually doing. So while indeed it can be a touchy subject, I really appreciate that these concerns and these design standards and principles are put forward and laid right out so that we can understand the thought process that is going into the AI tools that we are integrating into our systems. And then we can think about how we're using those AI tools in the systems that we in turn our building.

0:00All right, just in case we got mired in the mud on answering some of these fill in the blanks, let's go through them together. While blank blank includes a wide variety of technologies in the skill, we looked mainly at one specific subset of that larger concept. Of course, that is going to be artificial intelligence. All right, AI includes tools like machine learning, deep learning and the subset of AI, which creates images, audio and video, or blank AI, which is generative. Let's see, while data analytics and business intelligence looks backward and data to make extrapolations on future potential outcomes, well, that was a mouthful. Machine learning uses a wider array of data sets to make blank one word plural. That's going to be predictions.

1:02All right, machine learning has several ideal use cases, including replacing rule based if then systems, automation, personalization, and the ability to understand blank data. The nice thing about it is that it can understand unstructured unstructured data. Correct. In order for data to be considered high quality, it must be complete, unique, timely, valid, accurate, and be recorded in the same way every time, which is called consistency. And lastly, Google follows itself created set of principles when developing and utilizing AI. All of their tools include blank blank, which makes the function of the particular tool human understandable. This one you might not have remembered, but they refer to it as explainable AI. So explainable AI. And there we go. Hopefully you got them all right as well.

2:09I hope this has been informative for you, and I'd like to thank you for viewing .