today, chat, GPT, and why it's even bigger than you think.
My name is bill Russell. I'm a former CIO for a 16 hospital system and create, or this week health, a set of channels dedicated to keeping health it staff current. And engaged. We want to thank our show sponsors. We're investing in developing the next generation of health leaders, short test and artist site two great companies. Check them out at this week. health.com/today. Having a child with cancer is one of the most painful and difficult situations a family can face.In:
We believe in the generosity of our community. And we thank you in advance.
All right. I finally found a good article on chat GPT and it's by it's a blog article, actually, Josh Berson, B E R S I N. The title of it is understanding chat GBT. And why it's even bigger than you think. I'm going to do a lot of reading actually today. Cause it's, it's pretty good article. I'll throw in my anecdotes as we go along.
Ah, here we go. Everyone has an opinion about chat, GPT and AI. Engineers and entrepreneurs see it as a new frontier, a bold new world to invent products, services, and solutions, social scientists and journalists are worried with the one prominent New York times author as recline. Calling it an information warfare machine.
What has God wrought question mark. Uh, let me just say a front. I see enormous possibilities here. And as with all new technologies, we cannot fully predict the impact quite yet. There will be problems and failures, but the ultimate story is hooray. So he talks about what is chaff TBT? How can it.
Uh, how can chat GPT and similar technologies be used? And I think he closes up with, uh, where will the market go? So those are the three sections. Let's see, let's start. What is chatty, but to put it quite simply tech, this technology. And there are others. Like it is what is often called a language machine that uses statistics, reinforcement learning, and supervised learning to index words, phrases, and sentences.
While it has no real intelligence. It doesn't know what a word means. But it knows how does used it can very effectively answer questions, write articles, summarize information, and more engines like chat GPT are trained, programmed and reinforced to mimic writing styles, avoid certain types of conversations and learn from your questions. In other words,
The more advanced models can refine the answers. As you ask more questions and then store what is learned for others. While this is not a new idea. We've had chatbots for decades, including Siri, Alexa, Olivia Moore, the level of performance in chat GPT three. 3.5, which is the latest version. Is astounding. I've asked a questions, like what are the best practices for recruiting or how do you build a corporate training program? And it answered pretty well. Yes, the answers were quite.
Elementary and somewhat incorrect, but with training. They will clearly get better. And I did the same thing. I've asked it a bunch of questions. Got some really, um, I think good, good feedback. So chat GPT three was utilized by a program that we were using to actually write, uh, summaries for us on some articles and that kind of stuff.s the president of the U S in:
Such an a dollar believes 80% of code will automatically be generated. And I can write news articles, information, summaries, and more, which is what we used to for, I will say this. I had a problem with our website. I needed to Recode some stuff. I was having a challenge. And I gave chat, GPT the code.
And, uh, it corrected it. A corrected it for me. I put it back in at worked. So, um, you know, if you think about it, what it's done, it's consumed all of those sites where people ask questions. And, you know, then, then other programmers came in and answered the questions and then stars would elevate the.
Uh, correct. Answer to the top. It's consumed all that. So it has a pretty good idea how to write code for some. Uh, different languages. So a very interesting. Okay. Uh, one of the vendors I talked with last week and it goes on one of the vendors I talked with last week. It's using a derivative of GPT three to create automatic quizzes from courses and serve as a virtual teaching assistant. And that gets me to the potential use cases here.
Uh, let's see how can chats CPT and similar technologies be used before I get into the market. Let me talk about why I believe this will be enormous. These systems are trained and educated by Corpus. Database of information they index. The GPT three system has been trained on the internet and some highly validated datasets. So it can answer a question about almost anything that means it's kind of stupid.
You know, in a way, because the internet is a jumble of marketing self-promotion news and opinion. Honestly, I think we all have enough problems figuring out what is real, try searching for health information on your latest affliction. It's frightening, what you will find the Google competitor to chat to GP T3, which is rumored to be Sparrow was built with ethical rules from the start. According to my sources, it includes ideas like.
Do not give financial advice and do not discuss race or discriminate and do not give medical advice. I don't know yet if GP T3 has this level of ethics, but you can bet that open AI, the company who's building this and Microsoft, one of the biggest partners is working on it. So interestingly, I did talk to a physician who talked about one of his physicians because he's in a technology role. One of his physicians.
That, uh, use chat CPT to write a. Uh, referral complete with, um, uh, With citations and it did. And it did so, so effectively that it was submitted and, uh, the referral was approved. So it gives you an idea of the power of this thing. Again, it's not smart per se. But it understands how words are used and understands how to bring words together.
And write very coherent. Uh, thoughts around different things. Anyway. It goes on. So what I'm implying is that while conversation and language is important, some very erudite people, I won't mention names are actually kind of jerks. And that means that chatbots like Chiechi PT need refined deep content to really build industrial strength intelligence as okay. If the chat bot works pretty well, if you're using it to get past writer's block, but if you really want it to work reliably, you want it to source valid, deep and expansive domain data.
I guess an example would be Elon Musk over hyped, automatic driving software. I, for one don't want to drive or even be on the road with a bunch of cars that are 99% safe, even 99.9% safe. Isn't enough. Ditto here. If the Corpus of information is flawed. And the algorithms aren't constantly checked or checking for reliability. This thing could be a disinflation machine. And one of the most senior AI engineers.
I know, told me it's very likely the chat GPT will be biased simply because of the data it tends to consume. All right. So, uh, it's only as good as the data consumer is consuming data from the internet. We know that data from the internet is all over the board. Uh, plus it also learns from the questions you asked. So.
Uh, I did see a video of a person who thought they were being funny. And they asked GPT to solve a math problem, a simple math problem. Let's just say a thousand plus five. And I came back with 1,005 and then it said, no, that's incorrect. Do you have a different answer? And it came back with a different answer than 1,005.
Uh, and they played around with it some more. And again, they're training the machine to think a thousand plus five is not 1,005. So anyway. Uh, again, it goes on. Imagine, for example, if the Russians used GPT. Three to build a chat bot about United States government policy and pointed to the very conspiracy theory websites.
Uh, written. It seems to me that wouldn't be very hard to do. And if they put an American flag on it, many people would use it. So the source of information is important. AI engineers know this well, so they believe that more data is better. Open AI CEO, Sam Altman. Believes these systems will learn from invalid data, as long as the data sets.
Are large enough. While I understand that idea. I tend to believe the opposite. I believe. The most valuable uses of open AI. And business will be pointing the system at refined smaller, validated, deep. Databases we trust. Uh, I'm going to disagree with him here, but anyway, Regardless, he's giving you an idea of how it works and how you train these kinds of things.
Uh, it goes on in the demos I've seen over the years. The most impressive solutions I've seen are those which focus on a single domain. Olivia, the AI chat bot developed by paradox is smart enough to screen, interview and hire McDonald's employees with amazing effectiveness. There is a vendor who built the chat bot for bank compliance that operates as a chief compliance officer. And it works very well. Imagine as I discuss in a podcast, if we built an AI that pointed to all our HR research and professional development, it would be a virtual Josh Berson.
And might even be smarter than I am. Uh, let's see, I saw a demo of a system last week that looked that took existing courseware and software engineering and data science, and automatically created quizzes. I virtual teaching assistant courses online. And even learning objectives, the kind of work typically takes a lot of cognitive effort by instructional designers and subject matter experts. If we point the AI towards our content.
We suddenly release it. To the world at scale. And we as experts or designers can train it behind the scenes. And finally imagine the hundreds of applications in business recruiting, onboarding sales, training, manufacturing, training, compliance, training, leadership, development, even personal and professional coach coaching.
If you focus the AI on the trust domain of content, most companies have oodles of this. It can solve the expertise delivery problem at scale. All right. So where will this go to market?
As with any new technology, the pioneers often end up with arrows in their backs. So I'll chat GPT seems miraculous. We have to protect that innovators will advance, extend and refine this quickly. I would be willing to be the most, uh, bet that most VC firms. Are now writing blank checks to start ups in this area. So there's plenty of competition to come. My gut feel is that companies like open AI and Microsoft.
We'll likely compete with many other players, Google, Oracle, Salesforce service now, Workday, et cetera. So every major vendor will bulk up on AI and machine learning expertise. If Microsoft builds open AI API APIs into Azure. Then thousands of innovators will build domain specific offerings, new products and creative solutions.
On that platform, but it's still too early to tell. And my guess is that industry specific and domain specific specific solutions will went out. Imagine the number of opportunities, spaces out there to consider leadership development, fitness coaching. Physi, uh, psychological counseling. Technical training, customer service and the list goes on and on.
And that's why as early as this market remains, I still believe the opportunity is enormous. Uh, I liked him this tack to the early days of mobile computing and he goes on next paragraph and I discussed this podcast. A lot of jobs will change. I just did an analysis of jobs immediately impacted by chats, GPT editors, reporters, analysts, customer service agents, QA engineers, et cetera.
And found that today. With about 10.3 million jobs open about 8%, 800,000 will be immediately impacted these jobs won't go away, but they will be upgraded and enhanced by the systems over time. And I can confirm that we actually replaced a writer on our staff with, uh, a variation of chat GPT. Um, and so yes, but not only that there's programmers, right? Editors, reporters, analysts, customer service agents, QA engineers.
And more than one person. Expert that I've talked to about this are talking about programmers. Going away, the ability to talk to a machine. And natural language and essentially have a kick-out code. That develops an application for you. Again, that's not going to do way with a lot of the aspects of programming.
But it does take us to a new level of, I don't know, customization of the code that we have. And, um, I don't know, just a lot of stuff to think about. A lot of interesting things. I did a lot of reading in this. Because I thought the article was very good. I thought it was very well-written. Uh, explanatory on what chat GPT is, gives you an idea of what its strengths are and its weaknesses are.
And some potential use cases. There is a reason why Microsoft has invested. Uh, let's see. Here's the here's the article, Microsoft invest billions more dollars in open AI. So a multi-year multi-billion dollar investment. Doesn't have the exact number. Bummer. But anyway. Uh, so Microsoft was one of the original investors and then came back and really threw more money at it.
Um, and they really do see it as their. Uh, competitor to Google and then some quite frankly, so it is an interesting search way of searching for things. It is also, I like his idea of, uh, creating a bunch of Azure API APIs. To tap into chat GPT 3.5. I think there's a lot of power in that and a lot of potential.
Even in healthcare.
That's all for today. If you know someone that might benefit from our channel. You could do us a great favor and shoot them a note. Let them know that you are listening to the show and that they can subscribe wherever they listen to podcasts, apple, Google, overcast, Spotify, Stitcher. You get the picture where everywhere. And if you can't find us there, go head to our website this week, health.com. We want to thank our channel sponsors who are investing in our mission to develop the next generation of health leaders, 📍 short test and artist site.
Check them out at this week. health.com/today. Thanks for listening. That's all for now.