EP 247: AI - Your New BFF or Tech Terror?

This is the Everyday AI show, the Everyday Podcast where we simplify AI and bring its power to your fingertips. Listen daily for practical advice to boost your career, business and everyday life.

Can artificial intelligence be both a good thing and a bad thing? Is AI your new bestie? Or could it be a tech tear? We're going to be talking about those things today and more on Everyday AI. What's going on y'? All? My name is Jordan Wilson. I'm the host of Everyday AI and we are your daily livestream podcast and free daily newsletter helping everyday people learn and leverage generative AI to grow your companies and to grow your careers. So if that is you, I think you're in for a treat today with our guest and and we're going to be talking about that. The highs and the lows, the pros and the cons. And if AI is your new bestie or attack terror, I'm excited for this one. But before we get started, we're going to start as we do every day with our AI news. And as a reminder, whether you're listening on the podcast or joining us live like Rolando from South Florida or Tara checking in from Nashville, Woozy from Kansas City, make sure to check out today's show notes as always for or more and on our website@your everydayai.com all right, let's get into the AI news for today. So first, OpenAI has released and updated its GPT4 Turbo with new features. All right, so OpenAI has officially launched and updated GPT4 Turbo with vision in the API and within the ChatGPT interface, featuring what the company is calling significant improvements over its predecessor. So some of the more technical updates now include the ability to analyze text in JSON format and execute function calls, enabling developers to automate tasks like sending emails, posting online and making purchases. Also, the knowledge cutoff date, which this is a big one, will reportedly be moved to December 2023, enhancing the models relevancy and accuracy. So big jump there from April 2023 to December 2023. Also, competition in the AI space is obviously intensifying with Anthropic's Cloud Opus 3 Claude 3 Opus surpassing GPT4 in some closed source model rankings, while Google's Gemini Pro closely follows suit. Speaking of Google's Gemini Pro, new updates there as well. So Google has introduced Gemini 1.5 Pro with some advanced capabilities. So Google's latest Update equips Gemini 1.5 Pro with the ability to process audio files directly, enable it to extract information without relying on written transcripts. So this announcement came part of Google, came as part of Google's Google Cloud Next 2024 event going on now. Google announced the public release of Gemini 1.4 1.5 Pro via its Vertex AI platform, which is where it's available now if you want to go play around with that. So Gemini 1.5 Pro supports processing prompts containing one hour of video, 11 hours of audio, 30,000 lines of code, or over 700,000 words in a single stream with that large 1 million token context window. All right, last but not least in AI news for the day, that's been twisting my tongue, but intel has unveiled its gaudy 3AI chip, a what they're calling a game changer in the semiconductor industry. So intel has just introduced its gaudy 3AI chip at the WEF at the World Economic Forum meeting in Switzerland. So the Gaudy three chip is apparently, according to intel, over twice as power efficient and 1.5 times faster. Faster than Nvidia's H100 GPU offering enhanced performance for AI models. So we'll see once the benchmarks come out. But if this is true, that's pretty big news from in from intel, obviously. Nvidia has announced its newer chip, so we'll see where kind of the benchmarks fall with all of these things. Expected availability for the new chip for customers is coming in the third quarter with collaborations from companies like Delta, Intel, Hubert Packard Enterprise and Super Micro. So intel is aiming for competitive pricing against Nvidia's latest chips, emphasizing open integrated network on chip and industry standards. All right, so exciting. So we had big model news today, big chip news. But I'm excited to talk AI now. Y' all like I like. One thing I love is, is, you know, bringing our, our live stream audience the ability to, you know, learn live from experts. So very excited to bring onto the show. Let's go ahead. There we go. So welcoming to the show, we have Sheetal Rishi, an AI transformation leader. Sheetal, thank you so much for joining the everyday AI show.

Perfect. Thank you for having me. Jordan. Nice to be here.

All right, so much going on in the world of AI, but tell us a little bit about your background.

Sure. Oh yeah, by the way, I loved the daily three minute update. Everything that's going on in the world of AI and I think you do a great job in picking up what really matters. I think the top of top news. So thank you, Jordan. That was fantastic. So my background, I've been in the industry for, I've been a professional for almost two decades. Feels like just yesterday. And I'm still learning in my journey. But I started my career at McKinsey, did a lot of business strategy transformation engagements across the globe, advising C suite of customers essentially, or what they could do to transform new business challenges, new opportunities, how do you grow and stuff like that. And then from there onwards I had an opportunity to work with financial institutions, Citibank for a couple of years, actually deploying a lot of my advisory to the bank itself, how to transform and build the effort. There was something called One City that was part of that. So a lot of fun, exciting learning experiences on what does it really take to deploy Some of your own advice. I think I'd like to say that to myself. And from there onwards, I'm sure you would have heard of Deputy. So IBM Watson was the first, or one of the first supercomputers, if you will, for the lack of a better word, that actually took on the challenge to defeat humans in the game of Jeopardy. If you haven't seen it, you must check it out. There's a lot of great YouTube videos about that particular episode. So I had an opportunity to work with IBMs, spent close to 7, 8 years using AI to actually solve for some of the most pressing needs of the world, if you will. And then a couple of years in the middle between with deploying AI, building the infrastructure itself at hpe. What does it take? So the announcement about intel, especially of relevance and of interest as well, to go back and figure out what is this chip going to do and how is it going to impact my current customer base, I think is, is what I'm really intrigued about.

Yeah, yeah. So I'll, I'll go ahead and say this out loud. Your, your background is extremely impressive. Right. So you worked with some of the largest companies in the world. You know, you mentioned, you know, IBM, HPE, Citibank, McKinsey. So, you know, I'm curious, you know, from someone that has a, a deep background, you know, working in, in artificial intelligence and in business, you know, in business in general, how have you seen kind of the business world change, you know, specifically since generative AI, Right, since, you know, traditional AI has been used for decades. But from your vantage point, how has the generative side really changed how businesses operate?

That's a great question, Jordan. And has it changed? The answer is absolutely yes. You know, there's a lot of boardroom conversations. I mean, the fact that this morning at 7:15 your time, I think you're talking about AI, should be very revealing that there is just a Lot of excitement, a lot of hype about AI itself. AI is not new. I think the cat is out of the bag. That's how I see it. We've already been using different forms of AI. Again, just to throw in some more terminologies here. Machine learning or deep learning is not relatively new. I think it's been around since 1950s and 60s. I wouldn't call it a well kept secret, but probably it was one of those things. It was accessible, but we did not have the infrastructure that is required to scale and use these technologies, you know, at larger scale. So just like, you know, what electricity did to the industry back in 1880s when Edison literally had the first light bulb moment. Right. He started off in, I think in 1882 in New York, there were about 50 buildings that were lit up. They were the first one. There's obviously there's a lot of excitement about it, about legacy itself. And there was also a lot of concerns, you know, what if people start getting electric shocks, if you will, what it can do, what's the capacity? I think we're in a similar revolution right now, if you think of it. So the boardroom conversations are definitely picking up on how can I use AI. And I'm using AI as a very generic, broader terminology. The technology capabilities have become extronomically different is what I can tell you. And I was listening to another podcast myself last evening and it was interesting to hear a comment. Again, there's just so many statistics around us which sometimes I wonder which one to follow. But I think the net of it was that the speed of capability advancement over the last 10 years is just phenomenal, that you and I can actually use AI ourselves as individuals without having to invest, you know, hundreds and thousands of dollars. So I think the capability itself has been democratized. It's available to larger audiences. So what will happen with that really depends on human creativity and human capacity at an enterprise level. I am seeing much wider adoption of where it was. A lot of still, you know, barriers to entry, if you will, from an infrastructure capability perspective. The amount of data integrity that's required in enterprise regulatory requirements, the skills of the humans itself are actually going to deploy it. There are those. But I think those hurdles are becoming smaller and smaller as we move along. So the next five to 10 years of how we do business, how we see business, you know, it could be financial, it could be healthcare, would be very, very different from what we've seen in the past, in my opinion.

Yeah, and Sheetal, you know, you kind of mentioned that yeah, AI goes back many decades, you know, back to the 50s and 60s and the expert boom of the, you know, the experts boom of the 80s. But, you know, something you mentioned there that, you know, hopefully we can, you know, talk about here quick is, you know, Watson, right? So, you know, working, you know, on the Watson team, I think, you know, that was at least even for me, that was kind of my first, you know, kind of introduction to AI, right? When you had an AI, a supercomputer, go on Jeopardy. And win, you know, maybe. How do you think the conversation has changed, right, in, in the past, you know, 12, 12 or 13 years since Watson, you know, went on Jeopardy. And I think that really opened a lot of people's eyes to what was capable with AI. So, you know, as someone who was involved on the Watson team for, for a few years, how do you think that the conversation has changed since that very pivotal moment?

Yeah, great question. And again, I think three things that stand out to me and I'll try to boil the ocean here. One, I think number one is Watson itself. So conversational systems, right? They're typically, you ask a question, you get an answer. And the way they were trained, we used to call them short tail, long tail question. Short tail was frequently asked question. What's your address, Jordan? That's a, you know, short tail question. So if I have a public record or if I have a record in my system of records and I can find Jordan and, and I can find your address, I'll pop that answer back at you. This is, hopefully you're not asking what is your address? Nobody else is asking for you. But in the world right now, where we are, if I ask the same question, I think technology has come smarter enough to actually triangulate, pick up pieces of information from different sources and compile or what is called as, generate an answer and feed it back into you. That is one big difference. I don't necessarily have to have question and answer page pairs versus, you know, I can give you a dump of data and the patterns are being recognized. Just as we, as you're building a sentence, we need to know where the verb goes, where the noun goes, where the subject and predicate goes. In a language, right. Technology does it for us, infuses the words and I think the accuracy rate is. If you haven't used Chad, if anybody has not used Chat GPT, please, guys, go and try it. I think, and again, I'm not active advocating for one technology, but there are many different versions of Chat GPT or Gen AI Conversational systems, go use it for yourself and you'll see how it has evolved. Right. Number two was the data literacy itself in organizations. So back in the days a lot of my conversations will begin with this is AI, this is ML, this is deep learning, this is how the neural networks run. I think the conversation has gotten much smarter now the enterprises employees executive across 360 degree board. I would say, you know the, the newbies definitely are way ahead. People who've been in the industry and the executives, their level and appetite and understanding of technology is much far superior than what it used to be. And part of this is I think driven by the fact that many hyperscalers, many education institutions, you know, Coursera, edX, MIT, they're offering courses that anybody can go indulge in. So I think plus enterprises are having structured programs about AI and I think it's the gamification of education industry in this space has completely up leveled the bar in terms of what I know about technology now versus what people generally knew about 10 years back in the days people had sort of heard about it widely. It was a child, you know, poster child of everybody. I wanted to know about it but I think this, the skills level, skills parameter in the industry has really evolved significantly. The third piece is back to the infrastructure side of the story. A lot of times I would go we'll build many PoCs, demos, prototypes. It would take probably a couple of days to a week or multiple weeks to build the demos of these AI inherited solutions which is fantastic. But we were really getting caught up in the infrastructure. Then bigger issues around how can I infuse data in real time. A certain information has changed how is going to that impact the model? And the model throws a different result. So those limitations I see are getting much better right now. So things that costed $10,000 per unit and again I'm saying unit as an example of storage, compute, network. The cost and the speed at which we could access them has probably is probably, I would say few hundred times cheaper. I'm probably going to throw a number out there. I think it's at least lower by thousands right now. So if anything that costs you $10,000 probably at $10 right now available out there. So there is huge transform exponential, you know, transformation going on. So imagine what could happen the next couple of years. I don't want to look out 10 years quite frankly because I think there's just so much unknown out there. But in the next couple of years I see the opportunities really enhance significantly.

Yeah and that's such a great analogy that you said, you know, the computer, you know, that used to cost maybe $10,000, might now be 10,000 and what that means for business transformation. And I think that has a lot of people both excited but also maybe nervous, right? Because if, if they're not, you know, adaptable enough, I think now more than ever companies are feeling like, oh, I might get passed up because it is so easy, you know, to implement, you know, data transformation, you know, company wide. So, you know, that begs the question right to our original topic here is is AI your your new bestie or your tech terrorist? Tech terror? Which I love that. But she tell like can it be both? Can AI like be your, your best friend and can it be a tech terror or does it depend on, you know, your, your leadership and your ability to, to implement?

Are you still running in circles trying to figure out how to actually grow.

Your business with AI?

Maybe your company has been tinkering with large language models for a year or more, but can't really get traction to find ROI on Genai. Hey, this is Jordan Wilson, host of this very podcast. Companies like Adobe, Microsoft and Nvidia have partnered with us because they trust our expertise in educating the masses around generative AI to get ahead. And some of the most innovative companies in the country hire us to help with their AI strategy and to train hundreds of their employees on how to use Gen AI. So whether you're looking for ChatGPT training for thousands or just need help building your front end AI strategy, you can partner with us too. Just like some of the biggest companies.

In the world do.

Go to your everydayai.com partner to get in contact with our team. Or you can just click on the partner section of our website will help you stop running in those AI circles and help get your team ahead and build a straight path to ROI on Gene.

Again, it's, I think there's a lot of debate in the industry right now that is AI going to replace humans? Let's face it, you know, I don't know the answer to that yet with, I cannot say it with 100% confidence. What will be the future of jobs looking like? But I do know that people who use AI versus who don't use AI, they're definitely at risk, right? So start incorporating AI in your, in your daily lives if you will. Not just for, you know, going on Netflix or Amazon prime movies and picking up, letting the computer tell you which, which topic you should watch, which movie you should watch or whatever episodes you want to watch, but use it to your advantage and use it for your own efficiency improvement. For example. And most of the organization enterprises, I urge everyone out there that because employees will use it right and employees can use it, they're pretty vulnerable in my opinion. I mean start at home, your children especially for example in New York back in the day I think again back in the day it feels like, but I think when OpenAI ChatGPT came out first there was a lot of nervousness. There is going to be a lot of plagiarism. So kids should not be using AI, but I don't think that's the way to fight with the technology. We don't have to, it's always going to be human and AI. Think about a scenario when you can use a machine to actually do the task for you, but you still own the decision making capability that what task and what outcomes are expected out of you, I think then it's your bestie. But if you still continue to operate in a place where you're trying to compete with AI and memorize things and try to win over it, that's not going to happen. So if your tasks, if your daily work involves lesser of judgment, lesser of rules based decision making, I would really encourage everyone to start elevating your own profile, your own value add that you bring in. Because in my opinion the tasks that are more rules based, data entry type jobs and they're just a task in itself, they're not interconnected to one task after the other probably are going to be at the highest risk of being disrupted in the near future. So that's where that class of people is definitely a terror now is it? You use it as your bestie or as you know someone is going to replace you. Totally depends on you, how you're going to uptick yourself, invest in yourself, prioritize yourself is what I encourage everyone independent small business owners, you know, individual contributors and enterprises, large scale business owners, whoever you are, it's for everyone. The good news about AI is it's just like Lexity. You could do anything. You could turn on, you could turn on a computer, you could turn on a tv, you can turn on a light bulb, you could do what you want, you know what you, what your needs are. I don't think anyone, you should need to wait for somebody else to come and tell you what your needs are. If you can identify where, where you want to play. There's a lot of AI policy, AI governance training, AI securing AI cyber security is going to be a big hot topic to go after depending on what skill sets you have. Go and invest in your technical skills, on your soft skills on a domain knowledge, pick up an industry, banking, healthcare, everybody's AI will look different from the others. Quite frankly, I don't think there's going to be one solution that fits everybody. So pick up your battle, operate, you know, elevate your game, your skills in, in your specific area that you want to go after. And I think you're going to be fine. AI will be your best bestie.

So I love what you said there Sheetal. Like if you're, if you're trying to compete against AI, that is when it will or could become a tech terror. So one thing that I'm curious about, you know, you have a very, you know, deep background in working in enterprise. You know, if you, if you're just joining us in the middle of the show, maybe you missed your background there, but working at Citi, working at hpe, working at IBM. There was a study recently, I believe From Cisco, a 2024 study that said still 27% of enterprises are banning generative AI use. With that in mind, do you think, just speaking generally here, do you think that those type of companies, is AI going to become a terror for them or, you know, at least in the foreseeable future, is it going to be hard for enterprises to compete if they're not using generative AI from top to bottom?

Again, this is me speaking. I do not represent my enterprises and organization. I work for personally the pace at which I'm seeing technology and how useful it is, I really think if anybody decides not to invest will be left behind. I'm also a little surprised and my eyes, my head rolls when I hear that number from Cisco and I'm not saying that's a wrong report, but I'm really thinking there's probably more laggards, a lot of conversations. I definitely can speak with experience, firsthand experience of that about AI. I do understand AI is not click to deploy. Right. It requires, I think building solutions is probably going to be faster than driving adoption. So enterprises are really looking to, I think, understand how you apply. How do you drive change management in your own enterprises, legacy enterprises especially. And I think that's one of the reasons why you see startups in almost every sector because they're able to. There's not a lot of legacy. So there's a famous book that recently came out, the Fusion Strategy. If you haven't checked it out, people might want to read it, but the company really dramatically talks about automotive sector, multiple sectors actually you Know for example, I mean the way Tesla came up and about and they were able to, you know, disrupt the traditional GMs and Ford motors of the world, you know, and, and I think with all due respect, the legacy automotive companies are still trying to catch up with how the software automaker has actually turned the entire industry around how to introduce the first electric vehicle in the market. So there is definitely an internal resistance both from a skills infrastructure and what to prioritize and the investments required to trans legacy company. So it's a matter of time in my opinion that how when these legacy companies will turn around. So I don't think there will be any company that is not an AI company in the near future. It's just a matter of time.

Wow, that's. That that really makes you think about, you know, the future of business and, and where you know us as, as business leaders and you know, leaders in our organization should be focusing on. I, I think that's a, some very, you know, you know, powerful words with someone from a lot, with a lot of experience. You know, you're kind of mentioning, you know, different, you know, you talked about the book and we'll, you know, make sure to link that in our newsletter today recapping today's show. But maybe are there any sheetal many maybe any good use cases that, that you've seen recently? Because that's I think something people are always interested about. Right. Like people want first they want to be able to win back their time first and foremost and then they start to worry about like hey, what are some good use cases from your vantage point? Maybe what's a good use case that you've seen where companies have maybe actually correctly leveraged AI and it became their, their bestie.

Yeah, Couple of things I think many of us who are in corporate sector probably spend hours and hours of attending meetings and then hours and hours of synthesizing what we heard, what the priorities are and then hours and hours on tracking what should be done and then hours and hours on communicating making that the entire project end to end. Everybody is on power. So I'm connecting multiple steps here. Right. So the use multiple use cases and rituals could be please use AI and again I would encourage you to use AI for people in the, in the companies use AI that's approved by your enterprise. So GitHub Copilot is one of the most fanciest thing that I think all engineers are wishing and craving. Man, I wish I had it back in my days. So those are great examples of how it's helping in your code development itself and collaboration on code. So use incorporate gen AI for your own efficiencies. If you're going into a meeting, please urge your enterprise organization essentially to help you leverage a gen AI technology. And I think it's a CIO and CTO's nightmare right now because if you don't allow, people are still going to be able to use it because they're out there. You know it is available so easily so use it for your own recording meetings, documenting your meetings and sending minutes out, synthesizing what matters, what doesn't matter. It's a huge time saver. You could do it much, much faster. But again I don't want to encourage go wild wild west. Please use it as approved by your enterprise because there's just so much of enterprise data and the issues around data security privacy. I think we're still going through them. So use what you are allowed for. If you're not allowed for anything, definitely talk to your CIO and CTO organizations who are I'm sure are working around the clock to make sure that technology reaches you faster than you think. The other thing that we incorporated for my own work was a visualization dashboard, essentially pulling the brash if there are multiple stakeholders involved in your own engagement. So you know exactly where you play but you also know people who come before you and after you from an end to end supply chain. How are they doing so online live workflows if you will with a very simple visual that is something I think every business unit must look to invest and build it. Very simple. Again, I'm not here to talk about vendors and products, but Power BI for Microsoft has been my most favorite. Most recently Google has come up with a lot of feature functionalities that lets you build these visual. You don't need really to know the SQL code to actually build a program and then go deploy it and then the governance becomes a nightmare. But I think these tools are much simpler to use. Please incorporate that in your life. Two tools. Anything that helps you document stuff and anything helps you visualize your stuff that you can then democratize and collaborate with your peers across the world. Try to build start there. You know, I think.

Wow. So y'.

All.

I don't know if anyone else is is feeling like I am right now but I have so many notes here like so many great takeaways. I actually think I have more notes from the conversation than I did preparing for today's conversation. But you know, so she told we've talked about a lot but as we Wrap up. Because we've gone over how AI has changed over the decades. Some great enterprise implementation tips. AI optimization versus job loss, practical use cases. There's been so much good, you know, so much good information from our conversation. But you know, as we wrap up, maybe what's your one main takeaway that you can suggest out there to business leaders? Whether it is, you know, a small business CEO, whether it's someone working in an enterprise right now in charge of AI implementation. What's that one thing that maybe companies should be doing to make sure that AI is actually their best friend and doesn't become a terror?

It's a great question and puts me, I wonder because there's definitely more than one thing.

It's hard, right? It's hard.

It's a hard one. Yes. I really think, I think we humans are very, very smart. Innovation is distributed in our times today, right? It's not in the hands of one or two people telling us what to do. So I think we should enable innovation, we should enable creativity and how. What I mean by that is that as an enterprise leader, I think I'm a big fan of learning and this is a team sport. So if you know something and I don't know something, we're not going to be able to move the needle here. So I think enterprise wide, it's a revolution. So invest in learning. Give the tools to your employees so they can come back to you with solutions. So I heard a comment that's interesting, distracted with that, but yes, absolutely. So 90s man. Come on, who does that in these days? So invest, educate. I think investing in enterprise education will help you as an enterprise leader so people are not distracted. There's just so much being thrown at us at astronomical space. So if we can create structured, you know, just like in an academic institution, these are structured environments to promote education. I feel like, I think within enterprise itself, the learning function has to be very well thought through. It'll obviously require inputs from what the company strategy is, who you want to be, what kind of skills you want to develop, what do you want to, what do you want to be known for, right? So it should infuse, it should boil down to that learning catalog and then that should get passed on to different roles in the organization. AI will be embedded in each and everything that people are doing. Data scientists, engineers, architects, business leaders, you know, finance leaders, HR person. I just can't think of any function in an organization that will not, that does not have a use case for AI yet, in my opinion, you know, with Respectfully. So. So let's start with enabling our teams, our people, with the right tech that's relevant to our strategy.

I love that Sheetal saying that AI is definitely a team sport, right? And I think that sometimes people don't think of it that way. So thank you for joining us. Thank you for putting all of this into perspective and for sharing your insights on the Everyday AI show. We really appreciate your time.

Thank you for having me. It's great.

All right. And hey, as a quick reminder, there was literally so much great information there that Sheetal shared with us. So if you haven't already, make sure to go to your everydayai.com. we're going to be recapping everything that was said because maybe, like me, you were jotting down notes, but you still couldn't keep up with all the great insights. So make sure to go sign up for the newsletter. We'll have more there and we'll have more the rest of this week and every day. So please join us for more Everyday AI. Thanks, y'. All.

Sounds good. Thank you.

And that's a wrap for today's edition of Everyday AI. Thanks for joining us. If you enjoyed this episode, please subscribe and leave us a rating. It helps keep us going for a little more AI magic. Visit youreverydayai.com and sign up to our daily newsletter so you don't get left behind. Go break some barriers and we'll see you next time.