2021 Look-Ahead, and Bigger Use Cases for AI - Solving AI, January 2021

5:30PM Jan 2, 2021

Speakers:

Jon Arnold

Chris Fine

Keywords:

ai

technology

chris

people

problem

jon

data

patterns

world

solve

events

good

big

apply

pandemic

technologies

happening

case

topic

micro

Welcome to watch this space, the podcast about future of work. Every month, we bring you insider perspectives on how digital transformation, emerging technologies in generational change are shaping the future of work. We are two analog guys finding the groove for all of this in today's digital world. i'm Jon Arnold, and these trends are my focus as an independent technology analyst in my company, J Arnold and Associates.

And I'm Chris Fine. I'm an independent consultant and advisor specializing in topics like workplace technology, security, and IoT. My company is Integrative Technologies, LLC. Hi, Jon.

Hey, Chris. Hi. And we are back virtually, just the way we left off, but now we're in 2021. So it's time for a new year, it's time for season four of our watch this space podcast series. And for those who were with us last month, we left off with a bit of a year in review. And I know the session went kind of long, but you know, there was a lot to talk about. Right?

There was there was absolutely. And I might say hope everybody had a happy and safe holiday season and new year and you to Jon, your family. Yeah.

Thank you. Yep. So far, so good. And I think we've all i don't i think we've all come out of it. Okay, it's it's not like anything has drastically changed, you know, when the clock strikes midnight for 2021. But I think New Year time always gives us optimism to think about what could be, and the fact that we got a pandemic virus means that we think we're gonna get a control on things, at least it's a better looking scenario than we had before we had a vaccine. But obviously, things are not under control right now. And we don't know if there'll be a peaceful transition to power later this month. But I think there's plenty of things to feel optimistic about.

Time to look forward, I think.

Definitely, definitely. So on that note, today we are going to touch on one theme, for sure, and maybe one or two others. And we'll get to that in a moment here. First, I just want to say, a shout out to our listeners who've been with us. Probably not a whole lot have been with us from the very beginning. But as we go forward with our podcast, Chris, we're fine-tuning as we go. And the technology I think has kind of settled now for how we're doing our recordings and productions. I still have to add music, but that is coming, folks. It's the beginning of 2021. And we're getting this better all the time, so just bear with us, folks. And Chris, I think we've certainly got a lot to look forward to in terms of the topics we'll be looking at this year.

I think so, Jon, I think it's going to be a very interesting year. And certainly, hopefully, better, very different from 2020. But it's going to be interesting to see when sort of the world resumes, you know,

Yeah, on many levels, right. I know, we're talking mostly about work. And I think work from home is kind of old news at this point. And then we've touched on in recent episodes, this next phase of return to office, and how businesses are going to manage all of that. I will say some of the recent events I've I've participated in as an analyst, there's been quite a lot of talk with some of these updated collaboration platforms, where the focus is not just on the white collar knowledge worker, but for that remote workforce, the field level people, retail people who aren't sitting at desks all day long. And there's a lot going on with this technology that's making it more accessible to these various, you know, elements of the workplace that cannot be worked from home. So I think that's a positive development.

It's good to see the technology advancing to help these folks out, because as we've said many times, we can't really just think about the segment that is able to work from home all the time. Now, as you say, purely white collar jobs. So the more broadly, the technology can help folks who have a lot of different kinds of jobs. You know, think about some of the tools, for example, that are being applied to the frontline workers now in health care. That's all very good, the more the benefits of the technology can be spread out.

Yeah, I mean, I think we'll see a lot more of that, like anything else where there's utility, there's going to be adoption. And I think it's healthy that they're looking beyond the conventional, you know, the worlds where the Microsoft's the Cisco is live because they drive a lot of this stuff. But everyone realizes now that that's just one slice of the overall workforce. But also to you know, we talk about work from home. It's not just work, right? It's how we live our lives, and everything is going to be filtered more and more through a digital technology lens, and through the lens of AI, and I want, and that kind of will take me, will us really, but pick my next subject, a little bit closer to that spot. So when we think about what are we going to look forward to in 2021, I'd like to focus on one theme in particular, and we'll see where we go timewise today. So Chris, just before I launch into that, just any, I don't know, top line impressions you're taking as we go into 2021?

Well, I think the topic that you want to discuss which is AI, really needs to be on the plate. So I think we should launch into that first job. I mean, I always have observations, but there is in fact a time limit.

Sure, sure. Okay, so you know, AI is a can be a big, big word, it can be a scary word. And what I want to talk about where I am optimistic for 2021 is using AI for good. We in our, our bubble of collaboration technologies and workplace change all that, you know, we tend to filter this big technology trend through that space of how do we make people more productive? How do we improve customer service? All of those are legitimate applications. And, you know, when businesses want to solve a problem, they find the best technologies available. But the same technologies, especially AI, I view these as horizontal things right, where they have use cases everywhere, where workplace is just one application. So I want to think bigger. And I've written about this for no jitter.

Recently, AI, because it's so all encompassing. It's really not. I don't think it really has a lot of utility when it's solving the little things. You know, AI is big, it's complicated, for a reason. It's well suited for solving the big problems. And what I wrote about recently was there's no bigger problem in the world right now than Coronavirus. And it's a little bit like when you know, we went to the moon in the 60s, you know, there was no bigger challenge than being the first country to be able to do it. That was the hardest problem to solve. And they found a way - limited technology by our standards - but they found a way and they did it. And I think we're back in the exact same spot right now, Chris, where there's nothing more important than beating this virus, right?

No, absolutely. And not only that, Jon, it's the next pandemic. Exactly. One thing that this has taught us is that we need to rethink how we handle this threat in general, right?

Yeah, yeah. And so our ability to filter these technologies, first of all for good. And second of all, for understanding the problem set, and then working back from that and say, what's the best way to apply AI principles to that problem set? And we can all think intuitively about what we need to do, you know, wear masks, stay home, social distance, all that stuff, but it's still happening, obviously, so we're not all getting the message. But anyways, onto the AI piece.

One of the events I attended recently, it was from NEC. It was a global event. And one of the keynotes they had was Dr. Michio Kaku, who is very well known - his stock in trade is as a physicist at Columbia, I believe. But he's well known in terms of writing books. You know, he's really good at popularizing complex science, making it more accessible to folks like you and me. I've seen him before, and he's very good. And he picked this topic of saying how AI can be applied to help beat the Coronavirus and he's not talking about a vaccine or cure. But again, what's the problem set? And I think you Chris certainly would know, you know, when you're in your world, you know, it all stems from problem definition, right.

And understanding the data. Exactly,

So the starting point for this is okay, it's a big problem. What happened before? Well, when the Spanish Flu hit, right, that's the standard that we're measuring this against. People did not travel very much right not like they do today. Right? We really didn't have air travel there might have been you know, passenger ships and stuff but you know, trains and but not not a lot of cars yet. You know, people didn't, their worlds were pretty small. And so you know, if we had today's technology they probably wouldn't have helped as much, because everything was pretty localized, and they still had trouble fixing it. But today, think about how quickly and far and cheaply we all travel now, and the patterns of travel are just so much broader than we had back then, you know, a century ago.

So the starting point for his talk was simply - AI can be applied here, because it's really good at identifying patterns that humans cannot see. Or find. And that's true, right? Reading medical research, right? It's gonna, it's gonna solve diseases faster than we could ever have done on our own. So the idea is, not only can AI help us find these patterns better, but also more accurately. And those are really critical because these, you know, what we call the super spreader events are really what's allowing this pandemic to kind of escalate and persist. And there's always going to be isolated, you know, cases, but when you can stop community spread, that's, that's what we're after here. Right. And the key to that is if we can't control human behavior 100% - some other countries that have beaten it, that's exactly how they've done it. Right? I think Singapore, maybe Korea, South Korea, right? That's, that stops it cold. But in our more liberal societies, people have a lot more movement, and a lot more reasons to do what they want, right?

Well, it's interesting, Jon, I've seen, I've been reading about this, too, you're kind of seeing two approaches that are different, and have been successful in different ways to some degree and one, one we've taken, which has been not that successful. One is what you say, right? That the iron lockdown type of thing. And that has been societies like South Korea and Singapore and etc. But what you don't often hear and what other societies have done, like Iceland is a great example. There's a very good article in The New Yorker recently about how Iceland beat this thing. And essentially, what they did was completely data centric. It's exactly what you're saying. They didn't so much apply AI to it as an army of people in that field. But because it's a relatively small society, and relatively homogeneous, or a lot of genetic commonality helped them to but they actually weren't that what they did was they made it so that they could detect whether if there was any kind of event, outside of the rules or outside of the norm. And they could do micro containment, so they literally knew down to the person who might have been involved, and they could then isolate.

And that way, they didn't have to clamp down 100%. But they were able to chase down an infection really quickly, and very on sort of a micro cellular way. And if you think about it, that requires a certain compact between the citizens and the government, which you may or may not be able to obtain anywhere, everywhere. And also, they're relatively isolated. So it was easy for them to clamp off traveling. New Zealand pursued a similar, successful approach. But you're right, if you could apply better tools to it that would incorporate the concept of mobility, as well as just local cells of this, you'd probably be able to have less, you know, negative economic impact, at least, you know, and be able to keep people safer.

Mm hmm. Yeah. And so where this gets really interesting is because it's a global pandemic, the scale of this is all over the world. And as I said before, AI is really built for big, complicated problems. If it was just in a tiny place, like Iceland, or you know, Israel was, it was really good at this early on, because they have good technology, their small geography, you can control it. But when you've got a global population getting it and global travel happening, at least up to a point, you know, the permutations for spread are just infinite. So what he's getting at is, the trick here with AI is not so much that AI solves the problem. It's that the clues for solving the problem exists in the very technology we're already using. But we're not filtering the data from that tech those technologies effectively.

And in particular, there are two things he's talking about. One is social media, which is a very good barometer of our behaviors, right? Whether it's Facebook, Instagram, Twitter, people broadcast their lives now like never before, so it's very easy to see if someone's got something. If someone tests positive for Corona, it's very quick and easy to find out who their circle of friends are, where they are, when. So all of a sudden, it becomes much easier if you follow social media data to track those patterns. And then of course, at the micro level, we're all using smartphones. So all of the mobility data is there. Right? So it's just a matter of can you take these data sets, and mine them for the intelligence, again, detecting not just the patterns, but the anomalies in patterns that say, okay, this person is tested positive.

Now, again, you raise the issue of, of privacy, right? Wait a minute, Facebook data is private. So there are these issues of where I think now you start thinking about public private partnerships, that you may have to have to say, hey, if we have this data, we can use AI tools and machine learning to identify a where people are, in general, but also what their patterns are, when they are either positive testing or in contact with others who are now you can with that kind of detail of data, not only do you know what to do, but you can do it quick, right, because it's all real time. That's the magic of this, to keep super spreader events down, you know, to a minimum,

Well, you could also have identified more quickly the fact that super spreader events and exactly this type of event or that kind of event, we're crossing the spread. Because if you go back a few months ago, I think there was a lot of confusion as to exactly how this behaved. Right? So like, if you had been able to apply better data analytics, which is really what AI is, in this case, if you were able to do that, and identify conclusively say three months earlier, that it was transmitted by aerosol, not just droplets, you know, but through the air, it was really, it was really that contagious. And if you had been able to just you wouldn't have had to really probe anything that was private, I think, because to your point, there's so much data out there, from social media and everything else that if advertisers can micro target you when you give them permission to do that regularly, then why isn't public health another good use case for the same data? You know?

Exactly. So the metadata itself provides plenty of crumbs, right, to track people in their activities?

Well, one thing that's interesting about AI, this is just an editorial opinion. But with AI, you can use far less sensitive data to derive important conclusions, right? I mean, because you're looking at very broad patterns of data that and you know, if you're looking for, you could certainly think of plenty of dangerous uses for this. It's a very powerful tool, it's like a very powerful type of tool where it can do good, or it can do harm, you can apply it, you take the tool like a hammer, there's all kinds of different, you know, sizes of hammer, there's types of hammers, you couldn't really have the world we have today without hammers. But at the same time, they're not good if you turn them the wrong way. And so, you know, maybe we just need to try to find the good ways for this technology. Right?

Yeah. And the fact that, you know, a physicist is giving this talk, right, he's not doing this on behalf of a vendor. He's talking about, this is what the science is talking here, right, knowing what the problem set is and the risks we're facing. And the fact that it's a global issue. If this was like, you know, if it was only happening in America, it wouldn't be as powerful in application. But because it's global, the scale of human interaction behaviors on a global basis, is simply just too vast for humans to get their heads around. The data sets are just so, you know, it's again, it's just like trying to solve cancer or put a man on the moon, the calculations you have to do and track and do it in real time. We couldn't do this without AI.

And he gave a great example of this, using air travel patterns. And he had this chart, graph that mapped out all these patterns of where people are going and coming from, from these hotspots like in China, and where they came into the US and they were able to determine - I'm not sure quite how this all works, Chris - but they were able to demonstrate that 60% of all US infections, came through entry points to JFK and Newark airports. So that's a really effective, I think application to help you zone in where is this virus getting into our country. Well, if that's where the hotspots are, that's where you know, you've got to really be on top of what's happening. And obviously, again, when those super spreader events or carriers are coming all of a sudden you have Yes, like solving a, you know, Detective solving the case. Now you've got the clues where to go, and what to do.

Well, it's going to be interesting, Jon - I feel and we're not going to get to our second topic - but that's okay. I think, you know, you can resolve to do good. Coming out of this, right? There's a window probably where, you know, it's possible for the world to say, what what could we do better? And, you know, certainly one aspect that could be very important is the application of AI to this public health problem, and pandemics, but you need the will to, right? Anything economic, well, you need the political will, you need, you need the cultural social will to act on the data when you get it. Right. So let's just suppose we'd had AI, and it had told us clearly what the issue was, you would still have to have the public will, and the leadership to act on it, right? So to me, the two go together, right? Because good leadership, and public awareness and public will will help. In its best - as has historically been shown - to help apply tools to positive problems, you know, and to act on those problems. But if we can't get that done, then I think the technology by itself is really kind of speaking into the void, you know, exactly,

Yeah. And again, coming back to man on the moon, you know, there was a clear focus. We needed to be the first country to put a man on the moon. And they found a way to do it, because there was leadership to say, we got to do this, and they, and they got it done. And we're in the same spot here with this. Again, public private partnerships, if the social media platforms of the world recognize that, if we could turn all their efforts into saying, how do we use this data to help beat AI? If everyone's on the same page with this, and yes, if there is a plan and a will to do this, then absolutely AI can save the day, You know, but then you mentioned, you know, using this for good or evil, so to speak, Chris? Yeah, we're seeing all of this in the shadow of, you know, the the intelligence hacks that are happening all throughout the US government, you know, on scales we've never seen before. And forget about military warfare. I mean, cyber warfare is going to be, is going to be the way I think in this world. And this is certainly, you know, AI has a big role to play here. So in the right hands, it can be great. But you know, bad actors, you know, they can use it just as effectively in their own way too, right?

Well, Jon, I think it's just a classic example of watch this space, you know,

Yeah, totally agree. And, you know, I talked about our theme here, Chris, being we are analog guys. Offline, folks, Chris and I have very interesting talks about what we do with our vinyl record collections and reel to reel tapes, and all that analog stuff, which still we think has value. But anyways, where is the soul of technology and digital in this digital world? I think that's kind of what I want to get to as a takeaway for our audience. And I think I think this topic speaks to itself. I mean, technology can have a very positive soul, in my view, when it's used for the right reasons. And I just can't think of a bigger use case than than Corona right now.

I'm not sure I think the technology has a soul, I think it's a question of the soul that uses the technology. I mean, one of the classic failings of technology is to be taken without contexts, to just apply just to have this power, to have this tool. And not to try to have a context and ethical context around it. Context for social good, or whatever, I'm not sure how to describe it. But, you know, if you just invent something as powerful as AI, and don't think about the ramifications of it, it's tends not to go with a good direction. So it's all part the soul has to come from the people who use it, you know,

Agreed, agreed, Chris, you know, AI in particular, because, you know, it's easy to look at it as the big shiny object, the panacea for everything because it seems so omniscient and you know, so broad, that you don't know what to do with it. And no one, obviously, most people can't understand it. Because it is complicated. So it's very easy to put some blind faith into this stuff and just think, Oh, well, we'll throw an AI at it and it will solve the problem. Well, yeah, that could happen. But just as likely not, like you said, Chris, for the same reasons. If the human side of this doesn't take the reins and deploy it properly, etc., then it's just another technology.

Well, at least you have to have a framework around it. Right?

Yes, yes. Okay.

So we will see, right? I mean, I think we're gonna see both good and bad. My opinion. I think AI will drive some amazing breakthroughs. It already is, in many ways. But I think I think it's going to be, there's going to be some challenges too, for sure.

Yeah, that could be a next topic, by the way. You know, self driving cars is a good example of that, right? For good or worse. But, we got to go. So...

...yes, we do.

So, that brings us to the end of our time for today's episode, folks. So we'd like to thank you for listening. We hope you enjoy the podcast. This is Episode One of Season Four, and that you'll continue with us as we explore the future of work here on watch this space. Now, you can access all of our episodes at WWW dot watch this space dot tech, or wherever you subscribe to your podcasts. And if you like what we're doing, please leave us a review or rating wherever you subscribe. And that is my takeaway and wrap for now. So I'm Jon Arnold.

And I'm Chris Fine.

And we'll be back next month with another episode. So, thanks, everyone for listening.

Thanks, everybody, and we'll be back next month.