Watch Cognizant’s Bret Greenstein on Infinia ML’s Machine Meets World.

Cognizant’s Bret Greenstein on Data, Data, and (Don’t Forget) Data

Join Machine Meets World, Infinia ML’s ongoing conversation about AI

James Kotecki
Aug 11 · 24 min read

Episode Highlights

This week’s guest is Bret Greenstein, Cognizant’s SVP and Global Head of AI & Analytics.

“90% of the work is data preparation, getting access to it, putting it in forms that are useful, pipelining of data. That’s the really exciting part, because data has a massive potential energy.”

“All of this data outside of your enterprise combined with what you already know has massive potential. What people should be excited about is asking themselves and their teams how to unlock that potential. What would happen if I could combine social data with real-time data, with local event data and know something going on in my town, right now, that might shift inventories, or pricing, or staffing, or something that would drive a substantial change in my business?”

“I think in a few years the business leaders of tomorrow will be data-native thinkers. They’ll be comfortable with algorithms, and they’ll recognize that the best decisions are made when you consult with your intelligent systems, not consider them a threat.”

“Companies that delegate absolute authority for data science and AI . . . to a technologist [are] really missing the point of the responsibility of a business leader to ensure that systems behave without bias and that they reflect the values and the goals of your company.”

Machine Meets World is Infinia ML’s weekly interview show with AI leaders.

You can listen as a podcast (Apple, Google, Spotify, Stitcher) and email the show at mmw@infiniaml.com.

Image for post
Image for post

Audio + Transcript

James Kotecki:
Hi I am James Kotecki, and this is Machine Meets World, talking artificial intelligence live today with Cognizant’s own Bret Greenstein. I’m going to read this to get your title right, your full title, the SVP and Global Head of AI & Analytics at Cognizant. Bret, welcome to the show.

Bret Greenstein:
Thank you James, it’s great to be here.

James Kotecki:
Bret, I love that title, there’s quite a lot of important words in there. When people hear your title, what do they think you do?

Bret Greenstein:
I think most people think that I somehow deeply have my arms rolled up in technology, and models, and data science, and that it’s about the technology. But, most of my time is spent on transformational discussions with clients. How do I help them apply AI and data to drive a new outcome? I think people miss that, I think they think that AI’s about the tech. It’s really about the business.

James Kotecki:
So, what are some of those conversations, and how do they go? What are some of the biggest sticking points when you’re having conversations with business leaders about why AI can or can’t work for their business?

Bret Greenstein:
Well, it comes from two sides. One is, for companies that still think of it as a technology, they tend to delegate too much responsibility to the data scientists. And they ask questions like, “Can you find meaning in this data? Can you find meaning in that data?” When the real questions are, can you help me retain customers better? Can you help me drive customer loyalty, can you predict what people might buy? You need to focus the questions on the things that have real business outcomes, and challenge the data science team, but also the business teams that give requirements, to really raise the bar on expectations.

Bret Greenstein:
Those are the conversations that I have, is to help people connect the potential of data, and artificial intelligence, and machine learning, to outcomes that would drive real business change.

James Kotecki:
I think it’s interesting you use the phrase “people still think of it as a technology.” Which of course it is, but from a business perspective you have to think about it differently. Is there an analogy here that we can use to another kind of technology where it would be ridiculous for business leaders to think of it in the technical terms that people might be thinking about AI?

Bret Greenstein:
Well, it’s not hard to think back to the time where eCommerce and dot-coms became so popular, and people were spending a lot of time focused on HTML, instead of on business. Very quickly, we’ve dropped all that. You don’t stick an E or an I in front of every product name to make it sound cool and internet-like, you really have to think more about what is the business online, and that is really the discussion.

James Kotecki:
Well, obviously if you look at the name of the company above my head, we put an ML in our name. Our company started a couple years ago. But even when we started, we were having this conversation. How much longer is it going to be cool to say you’re an AI company, or an ML company? Are you starting to see that shift, the equivalent of the E or the I dropping away?

Bret Greenstein:
It’s coming, it’ll take some time. I think what’ll be really great is that data scientists and technologists will reclaim AI from the marketing teams, where it’ll go back to what it really should be, which is the advancement of the technology to enable business. And then, companies that use AI in their products and services aren’t going to have to spend as much time talking about the AI, they’ll talk about the capability.

Bret Greenstein:
I have a camera that I work with that has an AI algorithm in it that can detect when a person walks past it. They don’t market it as a camera with AI, they market it is as person detection because that’s what it does. And I think the more that people start to do that, the more that companies can focus on what they can do. And it is powered because of AI, but it’s not AI itself that they’re marketing.

James Kotecki:
Do you think AI as a concept will even really have significant meaning in five, ten years? Because there’s that old famous saying, and I can’t remember who it’s attributed to, but that the thing that we call AI is just whatever computers can’t do right now. Whatever the sexy, slightly out of reach or new thing, that’s what AI is. But of course, AI’s been around for decades. It’s been detecting financial fraud, for example, for many, many years. As you say, we’re not as impressed with that, or calling that AI, we just call it based on how it works.

James Kotecki:
So then, does AI still have this kind of sexy futuristic allure? Is it almost more useful as a target than something that actually exists today?

Bret Greenstein:
No, I think it will become, especially in hindsight as you look a few years now and then look back, to say it’ll be really the definition of systems that are trained, versus systems that are coded. Once you move into things that are trained, things that are learning, things that are evolving, versus things that are coded and deterministic, you end up in a whole different place. So I think it’s as different as when people move from Waterfall to Agile, or to object oriented programming.

Bret Greenstein:
There were certain step functions that people took that changed the way things work and the way we view things, and that’s really what’s occurring, I think, in AI and machine learning, is these learning based systems. They’re fed by data, they’re shaped by data scientists, and ultimately, they have to evolve and keep up with changes in business, where static coded systems don’t. And they need to be updated, and iterated over and over. You know, learning based systems can adapt better.

James Kotecki:
So then, what should business leaders actually be getting excited about when they have a business oriented conversation about AI and its potential?

Bret Greenstein:
They should be excited about the potential of data. I think what most people miss is that most AI projects are actually data projects, 90% of the work is data preparation, getting access to it, putting it in forms that are useful, pipelining of data. That’s the really exciting part, because data has a massive potential energy. Since more of it exists today than ever and will only grow, and new forms of data are becoming interesting … unstructured text, mobile, social, IOT, real-time, geo-spacial, there’s tons of forms of data. All of this data outside of your enterprise combined with what you already know has massive potential.

Bret Greenstein:
And what people should be excited about is asking themselves and their teams how to unlock that potential. What would happen if I could combine social data with real-time data, with local event data and know something going on in my town, right now, that might shift inventories, or pricing, or staffing, or something that would drive a substantial change in my business?

Bret Greenstein:
It’s not unlike the local store manager who ran a corner store. They knew all their customers, they knew what was going on in town, they knew when traffic was going to happen, and they shifted umbrellas to the front of the store when they thought it was going to rain. They knew this stuff, and they did it intuitively on one store. But now, companies do it at a global level, and they do it because of all that data that exists. And then bringing meaning to that data is a competitive advantage. Knowing that it might rain in 12 places means that you might shift inventory in those 12 places an hour from now, and then shift it somewhere else two hours from now.

Bret Greenstein:
That kind of thinking happens at a larger scale. It’s the power of the local knowledge of that corner store, but applied at a global level.

James Kotecki:
So, getting into some specifics if you can, how are you seeing this actually make money for people today? Because obviously you can talk about data as the linchpin, it obviously is. It’s also obvious the reason that so many companies aren’t able to do this, because they don’t have access to the data that they want, or they don’t have access to the data that they think they have access to. Or, getting that data is harder, and putting it into the right format is extremely difficult. You mentioned it takes 90% of the time, in many cases. So where is this technology, and this process I suppose, if you want to put it that way, being best applied today, in your opinion?

Bret Greenstein:
Yeah. Best applied, or most applied. So there’s a lot of companies who are doing work using machine learning and data to help drive productivity improvements, take out cost, drive customer sat through things like chat bots, and forecast engines, and things like that. Those are amazingly valuable and accessible types of outcomes.

Bret Greenstein:
But the more advanced companies in almost every industry, banking, healthcare, life sciences, insurance, across the board, they are using it to create value. They are using it to drive customer loyalty by understanding people’s behaviors, and needs, and wants better. They’re using it to drive revenue by decreasing churn, and providing the right offers at the right time. They’re using it to do underwriting better, so that they can offer the right rate at the optimal risk to maximize customer acceptability of an offer, and minimizing risk to the insurance firm.

Bret Greenstein:
So there’s all these interesting use cases that I think are about creating value, driving up revenue, and increasing competitive advantage. Everyone is focused on, I’ll call them the low-hanging fruit, of productivity and lowering cost. But, it’s the higher level things that I think are so interesting.

James Kotecki:
You know, the technology is abundant, as you’re mentioning. The culture in companies to actually be able to adapt and implement this stuff may not be. So I wonder, where do you see that? How would you define the size of that gap? And, the importance of non-technological leadership, cultural, structural issues on actually doing this stuff, getting from a demo to actually working in reality.

Bret Greenstein:
I think it’s a massive gap. Culture, skills, awareness, understanding, these are massive gaps but they’re also being bridged really rapidly. It’s okay that it’s a big gap. The fact that people are building bridges and trying to fix it is really important.

Bret Greenstein:
If you look at the number of business leaders who are data-driven today, who understand that potential energy that we talked about of data, who are willing to apply machine learning, who consider that their job could be augmented by intelligence. Traditional leaders may not have been brought up, or trained, or have experience in that areas, but more and more new leaders do. What I see is an infusion of people who are data-driven. I can feel it in our own company as we continue to bring in new leaders who are very data-driven, whether it’s from marketing, or HR, other places, looking at retention, and conversion, and understanding customers. But, everyone of our clients does the same thing.

Bret Greenstein:
As they get in fresh thinking, the fresh thinkers who replace the previous leaders are inherently more data-driven. They’re more appreciative and respectful of the power of algorithms, they understand the management responsibility of managing AI and not just delegating it to technologists. I see that infusion happening across all the industry, so I’m actually excited, being encouraged, because I think in a few years the business leaders of tomorrow will be data-native thinkers. They’ll be comfortable with algorithms, and they’ll recognize that the best decisions are made when you consult with your intelligent systems, not consider them a threat. Once you consult with them, you can see what is likely to occur, what the best decisions to make are, and then test and validate your assumptions as data rolls in, and you can prove what’s working and what’s not.

James Kotecki:
You mentioned the consideration of AI as a threat. I think probably most conversations that maybe you might have around the Thanksgiving table, and certainly I have when you tell people that you do something in AI, go to that perceived fear, that sci-fi driven fear in many cases. Also, the job loss fear that comes from articles that are written all the time, about how much this technology’s going to replace people.

James Kotecki:
You used the word augmentation several times. What’s the right framework for business leaders to be thinking about here? And, is the term augmentation papering over the fact that yeah, some people are going to get replaced?

Bret Greenstein:
Jobs evolve, always. Every technology has evolved jobs. I think there’s going to be more jobs as a result of the implementation of data and AI technologies, inherently because they create more value. As soon as companies create value, they create economic good, it creates opportunity. Now, I do think that the nature of those jobs will change.

Bret Greenstein:
So, you asked two questions. One, is the augmentation a good description? And two, should people be threatened for their jobs? I think people should focus on continuous learning and continuing to evolve themselves in a world where analytics, and data, and machine learning and AI inform decisions. The more that they inform decisions, the more the nature of your role changes.

Bret Greenstein:
I’ll give you an example. It was not long before the pandemic hit that I was running a monthly forecast review and a monthly financial operations review. The reports would come to me, and then they would trickle up through the management chain, and people would see them weeks after I saw them. We changed, during the pandemic, to real-time reporting, real-time insights, and I get the same data that the CEO gets, at the same exact moment. So the nature of my job has to change, because I can’t focus on how to massage data to make it look good for someone, I have to know what’s going to happen before it happens. So we shifted our energy so that my role is more of being able to predict what will occur, so that when the report comes it’s not a surprise because it’s something predicted.

Bret Greenstein:
I think the nature of business leaders will be much more predictive going forward, with a real-time delivery on insights and results, and a predictive set of capabilities so you know what is likely to occur. And then, what decisions are the best decisions to make? So if I think about the nature of the roles, the roles change. You need to be data-driven, you need to embrace the power of those insights to predict what will occur, you need to know how to look at multiple models to figure out which one is most suitable for the conditions you’re in.

Bret Greenstein:
And, augmentation’s actually not strong enough a word. Informed, everyone should be informed by the data and the intelligent systems of their business so they know what is likely to occur. What happened?

James Kotecki:
I love that. Yeah, I love that because the semantics of it really change how you think about it. Informed is actually a simpler word, it’s a more powerful word, and it’s more intuitively correct. Do I want to be informed or not? Obviously I do want to be informed. Why would I ever not want to be?

Bret Greenstein:
Yeah. Yeah, if you’re driving in a car and it has a safety system on it, and it flashes up that you’re probably going to hit somebody if you don’t touch the brakes. And then, if you ignore that system long enough, it applies the brakes for you. That’s intuitive, why would you ever want to drive without that once you have it? It’s just better. But it’s not driving for you, at that point, it’s simply helping you to avoid a problem because it can see a little bit further ahead, calculate a little bit better, and give you a choice which you make or don’t make. And then, a safety system kicks in at the end.

James Kotecki:
What does concern you, or scare you, in the realm of AI?

Bret Greenstein:
Bias and poor data. I think what worries me the most is the companies that delegate absolute authority for data science and AI in systems that have impact on their business to a technologist is really missing the point of a responsibility of a business leader to ensure that systems behave without bias. And, that they reflect the values and the goals of your company.

Bret Greenstein:
The reason this is so important is that every system is trained on data, every AI system is trained on data. All data is historic by nature, it already occurred. Some of that data is very, very old. There’s systems being used, data training sets, that are decades old, that are being used, for example, to train data to determine whether a name is male or female. You can tell just from that statement how outdated that model is. Some of those I think are 18 years old. That leads to influence AI systems to be inherently biased.

Bret Greenstein:
Now, that’s not necessarily a bad thing, it’s just not a desired thing. People are also biased. So as leaders, when we hire people, we set values for our company, we ensure we measure people to those values, we constantly evolve them, so that we drive bias out of behavior in a way that might hurt our business for people. You have to do the same thing a learned system, with an AI based system, to make sure hiring, and pricing, and insurance underwriting, and all these processes, happen without bias.

Bret Greenstein:
So I worry that the mistakes companies make in this area are going to get overblown and people will be afraid, rather than recognizing this is no different than the bias that their human decision makers have always had. It’s simply that we now have a responsibility to make sure that we’re looking for it, we detect it, we fix it, and we make it better.

James Kotecki:
A business leader might throw up his hands at that, or her hands, and say, “Look, I don’t know the technology well enough. How am I supposed to detect bias in this stuff? I have to trust my data science team, I have no way of knowing this.” What would you say to that person?

Bret Greenstein:
When you hire human beings to do a job, and you look at the results of their job, you look for signs of bias. You look at people are hiring, or underwriting, or assigning risk in ways that are not representative of your values as a company, whether you discriminate or not. I think we expect business leaders to measure people that way. We need to expect business leaders to be able to do this as well, for the systems they implement.

Bret Greenstein:
If you implement a system and don’t even ask the questions … You don’t even have to understand the technology, but you have to ask the questions. Are we hiring with the diversity goals that are set our company for? Are we discriminating when we do X or Y, based on criteria that are not valid? Or not legal, or not part of our values of our company. These are discussions that business leaders have to have. Successful companies are addressing this now, it’s a topic in every boardroom, it’s a topic among every discussion I have with clients. It’s solvable. There are technologies to help detect when biased results come out, relative to what the desired goals are. But more importantly, there are questions business leaders can ask to make sure they’re being addressed.

James Kotecki:
Is this an issue that we’re going to need to leave to the business community? Or, should we be looking at major AI ML ethics regulation? Not necessarily thinking that it will happen imminently, but is this something that you think should happen or want to push for?

Bret Greenstein:
So, this is often discussed among government agencies. The conclusion that almost always comes out is we actually have laws in place already today, to avoid discrimination, to reduce bias, to protect people. Those laws need to be enforced as they are. There’s libel laws, there’s liability laws, there’s all kinds of various laws that apply as well for an AI based system. So most regulations start with making sure that the policies we have in place for people apply equally well whether it’s a system or a person.

Bret Greenstein:
But I do think that, over time, they’ll have to be increased regulation as well, to make sure that people understand their responsibilities. The same way that privacy regulations have come into play, whether it’s GDPR, CCPA, HIPAA, those were needed to be extremely clear about the protections people should expect, and the same thing will occur naturally here. I think it’ll be extensions of existing laws, for the most part. I don’t think these are really issues directly related only to AI. AI might scale the issue, it might amplify the issue, but it doesn’t change the responsibility that business leaders have always had to maintain ethical behavior and to avoid bias.

James Kotecki:
Is it going to require a new crop of legislators and regulators to actually get this done? You mentioned as new business leaders come in that have a data-driven mindset, things in companies start to shift. You don’t have to look too hard at the median age of a Senator, or watch some of those hearings where they’re grilling technology executives to varying degrees of success, to understand these folks don’t necessarily get enough to really do a good job of setting the groundwork here.

Bret Greenstein:
You asked me a hard question. I think there’s always a need for a refreshing of skills. I don’t think it has anything to do with age, I think it has to do with skills. For those Senators, and Congressmen, and policymakers, and business leaders, there are lots of forms of education and experience to be gained that will help them have a better context for setting these kinds of rules.

Bret Greenstein:
You know, I’m pulled into and involved in discussions with policymakers all the time, and I find the dialogue and discussion that you don’t see on television, the stuff that’s actually happening, to be fairly deep. There’s been some published papers recently by the US Government, specifically on regulations and policy for ethical behavior and bias with AI. I think it’s extremely well thought out materials. While it’s easy to look at a few examples and say, “They don’t understand technology,” that’s the show. But underneath it, I think the people responsible for these jobs take it seriously, and I think they’re focused on the continuing education.

Bret Greenstein:
Also, they reach out to industry. That’s the reason I got involved at all with any of these policy discussions, was simply that people reached out to industry also to get perspectives.

James Kotecki:
That’s refreshing to hear, and an optimistic note, I think, for those of us who do watch those hearings.

James Kotecki:
What inspires you about working in this field? What inspires you about AI? Does it tie back to, I don’t know, watching sci-fi as a kid? Were you into it as a kid?

Bret Greenstein:
Yeah, I was very into it as a kid. There’s a couple things. Obviously, the things I read and watched as a child were inspiring. The idea of systems that learn has always really just impacted me.

Bret Greenstein:
I had a teacher when I was very young, who came into school one morning and just said he saw a computer. He walked up to it, asked what his name was. He typed Ted and it said, “Hi, Ted.” I was blown away that a system could learn and remember something, and I just kept imagining what could happen. Then, you read the works of Issac Asimov and others, telling stories about the potential of robotics and AI, it just sounds amazing.

Bret Greenstein:
But, when you get into it, and start learning about the techniques and technologies, what really inspires me … That started me out, but what really inspires me is that we’re still very early in the technology. There are leaps and bounds made, every single day. And the people working in projects, and innovations, and AI, and data science, and machine learning, they continue to show cool things you couldn’t anticipate. So, that just inspires me, I love the projects that I get to see that our teams do with clients. I love the innovations that continually occur. We’re at a very early stage still, in the potential for data and AI.

James Kotecki:
Is there any cool emerging demo that you’ve seen recently, that you can talk about?

Bret Greenstein:
Yeah. There’s the obvious ones that a lot of people see, whether it’s Open AI and GPT3, being able to generate text, it’s really cool. Obviously, the self-driving capabilities, which are not full self-driving but approaching it in current cars are really, really cool.

Bret Greenstein:
But, we did a piece of work using Evolutionary AI at Cognizant. We showed it at AI Summit last year. What really blew me away was it’s a learning based system, we taught it how to play a game. In this case, it was a game called Flappy Bird. We taught it a video game, we’ve taught it other video games. But we don’t teach it by teaching it all the rules, we teach it by letting it experience the game and fail, or succeed, and teach itself. What it does is it adapts its own strategies, it’s creative. It finds strategies to be successful, improves those strategies over time, and learns very fast. This is the same type of work that was done by DeepMind to play against Go. They didn’t teach it every strategy masters ever learned, they just taught it how the game works, it learned its own strategies.

Bret Greenstein:
So we apply that approach, called evolutionary AI, in business. What’s really cool about it is it will learn a set of rules and conditions, like it learned to play that video game, and it learns faster than people, and it learns faster than a traditional deep learning system. But, what’s really cool is if you change the rules, if up means down, left means right, if you change the rules in some way, it re-learns very fast. And I think what really struck me about that is people don’t. People inherently do not learn very fast when the rules change. Which is why, in an industry or business, when the conditions of a business change, like the pandemic change, not everyone was able to pivot, and adapt, and learn new ways. Some did, some didn’t.

Bret Greenstein:
For companies that are truly adaptive, that have decision making that can learn, and when conditions change can adapt its strategies to find other strategies, that to me is game changing. I think if you’re a business leader of today and tomorrow, and you have systems that help you see changing conditions, and help you understand the changing rules and adapt to it, that’s a competitive advantage.

James Kotecki:
Can we draw broader lessons from the evolutionary AI video game example to how a more advanced AI, maybe something that’s approaching artificial general intelligence, I don’t know, would behavior out in the world?

James Kotecki:
So, let’s say you take a really smart computer, put it into a humanoid robot brain, like Westworld or something, and you set it out into the world. Well, life is not a video game with simple rules, right?

Bret Greenstein:
Right.

James Kotecki:
There are some rules, you think about laws as rules. But even laws aren’t always hard and fast. We all jaywalk once in a while, because we know there’s certain times that you can get away with it and nobody cares. We all go five miles over the speed limit, sometimes. I’m not admitting to anything for purposes of law enforcement watching this video, but let’s just say that most people do that.

James Kotecki:
When you take … Can you draw an analogy from what the computer’s able to do in that example that you gave, to how it might behavior in a much more open world, to use another video game term, environment?

Bret Greenstein:
I think that might be too big a leap. I love artificial general intelligence as an idea, I love the premise that writing, and movies, and pop culture have created for us as what the potential is for that, but I think the difference between the narrow use cases that exist today for most AI systems and artificial general intelligence, it is such a wide gap that I think it confuses people. So the idea that you could combine all these systems and create something that approximates an intelligent person, who would know if it walked on a street whether to jaywalk or not, is an interesting thought experiment, but it’s very, very far from where we are.

Bret Greenstein:
I’m not sure we’d even recognize it as it approaches, because it’s not going to be a moment. When systems get better at doing things, like today … 10 years ago, a system might be able to see a dog or cat, five years ago recognize a dog versus a cat, now it might recognize my cat. Next year, it’ll recognize animals it hasn’t seen. Whatever those advancements are, they happen gradually. When you add those to other systems, you approach the capabilities of some things, but it’s very, very hard to see when you would consider it to be truly intelligent.

Bret Greenstein:
I know that many people have talked about the Turing test. I forgot who said this, they had the coffee maker test. Could you set an AI into a house it’s never been in and ask it to make a cup of coffee? Could it figure out where the ingredients are, how the coffee maker works? These sound like things that intelligent systems or people should be able to do, but those are not really where people are focused the energy. Most of the energy goes into narrow use cases, natural language. Understanding of text for contracts, for example. Or, image systems to recognize defects in products. Or, an act in streets’ corners. Predictive algorithms to see when things are going to break. This is actually the dominant themes in the near term.

Bret Greenstein:
I’m excited for the potential, I love the thought of AGI, I just think it’s potentially a bit of a distraction for most business people because it’s so far from where we are.

James Kotecki:
You’re bringing up an interesting point, this is something that I was realizing recently. Long before we ever get whatever we’ll define as real AGI, we’ll get stuff that pretends to be AGI, or claims to be AGI, or someone has programmed to say is AGI, and it’ll be good enough to fool a certain percentage of people. We can do that now, there’s systems that can pass the Turing test now, by different definitions of how you run those tests. So I think you’re onto something, that maybe it’s better suited as a goal versus a specific threshold hold, once we cross it, we’ll be in this new world. It’s going to constantly be evolving.

Bret Greenstein:
It’s much more of a continuum. I like to go back to people as an analogy because I think it’s helpful to relate to. You know people who appear very smart, and as you asked questions you suddenly realize the limits of what they might know. They might be able to rattle off a quote or tell a story and it sounds really good, and then when you poke underneath it you realize they don’t actually know what’s behind that.

Bret Greenstein:
I think that’s the trouble with most of these systems, is once you start to really inspect them, really test the limits, you realize what they know and don’t know. Self-driving systems are easily defeated with electrical tape over speed limit signs, just to change the speed limit, the car drives incorrectly. Most systems are fooled by even … The tools to generate text can easily be questioned to a point where they start giving ridiculous answers.

Bret Greenstein:
So I think we’re expecting a bit too much from these systems today, and I think the focus on practical applications, while it may not be as science fiction-y, exciting, shiny as what you describe, I think it’s so powerful and impactful to the way work and business, and the future of work, is being done that we can spend the next decade building up these systems, improving business, and having a huge impact on the economy, on society, on businesses, on all of us.

James Kotecki:
Well Bret, you mentioned that you asked some people questions, and it turns out that they don’t have a lot going on under the surface. Your conversation was exactly the opposite, we asked you a bunch of questions, and it turns out you had some great insights for us. So Bret Greenstein of Cognizant, thank you so much for joining us today on Machine Meets World.

Bret Greenstein:
Thank you so much, James.

James Kotecki:
I am James Kotecki, your host. This has been Machine Meets World, a live production of Infinia ML. You can get this as a podcast, you can see it on YouTube, you can just share it, like it, tweet it, whatever you want to do. Thank you so much for being a member of the audience, thank you so much for watching, and we will see you next time. This has been what happens when Machine Meets World.

Image for post
Image for post

Originally published at https://infiniaml.com on August 11, 2020.

Machine Meets World from Infinia ML

Weekly Interviews with AI Leaders

Medium is an open platform where 170 million readers come to find insightful and dynamic thinking. Here, expert and undiscovered voices alike dive into the heart of any topic and bring new ideas to the surface. Learn more

Follow the writers, publications, and topics that matter to you, and you’ll see them on your homepage and in your inbox. Explore

If you have a story to tell, knowledge to share, or a perspective to offer — welcome home. It’s easy and free to post your thinking on any topic. Write on Medium

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store