Chet Kapoor (0:24):
Welcome back to the Inspired Execution podcast. Each episode shares the experience and learnings of a world-class leader on their journey to success. The guests on this podcast are bold, brilliant, and not afraid to change. As you navigate your own path, we hope you feel inspired by these stories, lessons learned, and the vision of the future.

We are super excited to have Jeetu Patel back on the podcast for a conversation about working asynchronously. Jeetu heads up security and collaboration at Cisco. He previously led Box through its evolution from a single product company to a platform company, quadrupling their revenue to over 700 million. Though his childhood dream was to be a flight attendant for Pan Am, he's pretty happy with the direction his journey has taken. You're going to love our discussion about creating connections in a hybrid workplace, the impact AI will have on how we work and so much more.

Hey, Jeetu, welcome to the podcast.

Jeetu Patel (1:24):
Chet, thank you for having me, man. It's good to see you.

Chet (1:27):
We've known each other for many years and this is your second time for the podcast. It is awesome to have you here. So, the whole world is talking about AI, which is very relevant to our topic today, working async. In a recent interview you spoke about the benefits of AI, but also mentioned the biggest risk is humans with bad intentions. What do you think about the strengths of a tool like GPT for productivity while balancing the unknowns? I mean, as a GM for WebEx, you're thinking about this all the time.

Jeetu (2:01):
Yeah. Firstly, look, I think if you were to look at major platform shifts that have occurred during our lifetime, Chet, you and me, you see a major move to the PC, client server, then to the internet and SaaS, and now there's going to be a major shift with AI. And each one of these transitions that happened, there were really good things that the technology provided and then there were really bad things that happened. Even with things like mobile, if you ask my 12 year old daughter, she thinks it's great. I think it's terrible that she spends so much time on the phone. The way that we think about AI is there's the biggest innovation that's happened right now over the course of the past few months since November 30th when ChatGPT came out, was natural language, and the understanding and the ability for a machine to do reasoning and all of that.

But I do think there's a lot of other areas where AI has actually been doing pretty well as well. So, for example, we've done a ton of innovation on audio intelligence, sound and acoustics engineering, and how do you go out and translate speech to text, and text to different languages, and all of those pieces actually make a huge amount of delta and how people can use technology. The second area that there's a huge amount of innovation that's happened with us is on video intelligence. You have six people in a room, but they're in a very far distant angle. You can't really see them. Can you actually automatically detect humans and get a focused view of that individual and make sure that you have a grid view just like we see right now? But with six people in a conference room, they're all sitting next to each other. Echo cancellation and there's a bunch of these things that have actually been small little ways that we've been chipping away at it.

The nice part about AI now is the compounding effect of all of these things, one on top of each other, has a stacking effect that's really, really cool. So, we start everything in WebEx with speech. But when you get from speech to text and then text to reasoning, the quality of reasoning gets better when you actually have better background noise removal that can then be converted into text. And then I think the big unlock that's happening with AI is not in the model, but the proprietary data sets that you have. So, imagine in a few months, in a few quarters you happen to be late to work one day and you got stuck in traffic, something happened, you couldn't make it on time, and so you got in at noon instead of 9:00 AM. You missed about six meetings and you're an important guy, you just go to WebEx and you say, "Hey, what meetings did I miss and was there any action items for me?" And it'll just tell you those action items.

That's the kind of magic that I think you'll be able to get going over here. And that's because of proprietary data sets that will enhance the model. The models are going to become much more of a converged commodity, where the Bard model from the ChatGPT model won't be that different over time. And if you give them both exactly the same dataset, the responses and the outputs you get will be kind of similar. But when you actually have one model that actually gets a little bit more data, you will see a step function improvement in the output.

Chet (5:03):
Yeah. No, I agree. It's funny, I was doing the shimmy when you were talking about it's the dataset, because we were just having this conversation. We recently got into the realtime AI space from a data stacks. We're doing a realtime AI platform, which is essentially bring the models to the data. And the tagline we were having a discussion about was, it's about the data, not just the model. And so, we're 100% in agreement, if I understood this correctly, everything that you talked about from a speech and video perspective and from a sound perspective is all about predictive AI. And one of the things that you will have to land up doing in your business, is to get more into the generative AI aspects of things, and make sure that the users of your product actually will not understand, you will blur those differences.

Jeetu (5:54):
What's super exciting about generative AI is the days of the mouse being the primary interface are numbered. I think what's going to happen is you're going to see many more command prompts with natural language, with bidirectional communication with the computer, where you ask a question, gives you an answer, gives you a suggestion. You build on top of that and then one thing compounds with the other and that's the interface. And then when you layer on top of that speech, over time, you'll actually see that you will be talking to a computer just like you talk to a human being. And it'll all be in this very natural sequence of events that occur.

So, you're right that the generative AI part and the predictive model will start to blur. I don't think there'll be these distinct things that happen. And frankly, for the user, it shouldn't matter. For the user what should matter is I have a question and I want an answer, and I want to make sure that AI can augment my understanding as a human. And I do believe that AI can go in multiple different ways. You can actually have AI fully automate a human out or you can have an AI model actually augment the human knowledge base and create original insights beyond what the human corpus of knowledge has within the world. And I think the latter is going to be where we will try to push the market.

Chet (7:17):
But you are in the automation business, you're to make me more productive. That's your goal of your product. And you are going to get really good at your game. And you are going to say what I'd get done in 10 hours, you're going to try to give me over a period of time to do it in two hours, and do it in one hour, in shorter period of time. So, I'm super pumped up about this.

Jeetu (7:39):
I think what the most underestimated part of AI right now is the original insights that an AI model will be able to add to the corpus of human knowledge that already exists, that I think people are grossly underestimating at this point.

Chet (7:53):
100% agree with you, because I think generative AI gives you context that you don't normally have. And it gets better because you give it context, and it gives you context back, and that interaction just makes it really awesome. I think of that as me just being a more productive leader. And hence, I put it in that bucket, but I think it's a very fair point.

Jeetu (8:12):
I do think it would be naive of us to think that the downsides are marginal and trivial, because here's what could happen. Let me give you the bear case of the scenario. Bear case of the scenario is every job could in fact get replaced quite dramatically with automation, starting from the creative jobs. And when that does happen, you will see a lot of displacement of labor. But I think that's what we have to actually be very careful of is when those jobs do get displaced, we have to make sure that there's policy, regulation, government assistance, private sector assistance in ensuring that that displacement does not create human suffering beyond a certain point, so that there's not a concentration of wealth that gets even more exacerbated as a result of this.

Chet (9:00):
And that's why I believe this is different for all of us. We've been through the consumer tech revolutions, through mobile, and through the internet. And we've been through the B2B or I should say the industry tech revolutions, where you're doing client server and things like that. And this time around we will find that policy and governance will have to play a role, not when things have been implemented, but earlier on. And so, we will actually have to do this in parallel almost. Take a step and then slow down, take another step and then slow down. What do you think? You think hybrid work is here to stay?

Jeetu (9:39):
Yeah, I actually believe that it would be. Regardless of what I do for a living, I'd say even if you keep that aside, I think it would be a sad day for humans if we tried to regress back to the pre-COVID world where everyone had to be in the office all the time. And let me tell you why I think that would be sad. There's this notion that opportunity is very unevenly distributed throughout the planet. You and I know this because of the background of where we came from, but I don't think human potential is that unevenly distributed. In fact, there's smart people anywhere in the world you go. And it always surprises me that the only difference between a really smart person in a village in Bangladesh and me is that I had a platform and that person didn't.

And so, in my mind, if you can afford that person in a village in Bangladesh, the ability to participate in a global economy and benefit from the economic opportunity that comes from that in the same way that someone in the heart of Silicon Valley can, I think you've just made the world a better place. By having geography become the limiter, I actually think it's counterintuitive to me on why that's good for society at large. And so, the next generation of leader in my mind should be one that knows how to build deep, trusted relationships without having met someone in person. You and I have known each other for, I don't know, eight, nine years, 10 years now. I don't think you and I have ever met in person.

Chet (11:07):
Correct. Correct. Still a lot of respect for each other.

Jeetu (11:11):
If we meet in person, that would be fantastic and I'm sure there'd be other things that would come out of that dialogue if we met in person. But that is not necessary for us to feel like we have a close bonded relationship. I have actually transcended three jobs, you've transcended three jobs, and across all of those jobs we've kept in touch. I kept in touch with you when you were at Apigee. I kept in touch with you when you were at Google, then I kept in touch with you when you were at DataStax, and vice versa for me in my three jobs.

And that I think is the skill that people will need to build. And yes, I think hybrid work is here to stay. There'll be some companies that will try to mandate, but I don't think that's going to work with 100% mandate because the best people are not going to want to be told where and how to work. Now, by the way, I don't think 100% remote is a great idea either. I do think that there are times when people coming together is a fantastic thing.

Chet (12:03):
One of the things I struggle with as a distributed company where we have a great office right here in Santa Clara, but it's not a place for work, it's a place for collaboration. So, people from all over our database team is here. All over the world, they come in, they're here for three days. It's very inviting. We have a theater, we have everything catered. And so, it's just a great environment for people to collaborate. One of the things that I miss is how do you read and connect with people on a screen. We've not cracked the code. And if you have any ideas, I'm happy to learn.

Jeetu (12:36):
Well, I'll give you a couple things that we are working on and a couple things actually are human elements, not technology elements. And so, one of them, why is it that you actually have a hard time on video connecting with people than you do in person? The big reason is the absence of wander time that's there. So, video is a very transactional means. I'll give you an example. If you have a 30-minute meeting that's back to back stacked with nine of the 30-minute meetings, and if in the first meeting you stall by about at 22 minutes where there's like a seven-second lull, I guarantee you in that meeting someone will say, "Any other questions or should we give back eight minutes to all of you?"

Now, imagine if you and I are at dinner and we have a seven-second lull, what I would do is immediately pivot the conversation to, "Hey, Chet, how old are your daughters now? What are they doing these days? What do you do for fun?" And what that does is it creates context, and that context creates familiarity, and that familiarity creates trust. And that trust creates a comfort zone for making sure that you can engage in conflict without actually taking it personally. And that I think is the missing element in video. And that's not a technology limitation. It is a human conditioning where we seem to think that small talk is a waste of time. Whereas, what we don't realize is icebreakers are in fact a way for people to create more of a trusted bond. And so, one of the things that we had done during the pandemic is we acquired a company called Slido.

And Slido allows you to go out and create these icebreakers and engage in an audience where it feels like you're having a conversation. And that's a pretty important dimension to keep in mind. The second big area that's hugely negative on the not in person is time zones. What do you do with time zones? And try to go out and schedule a meeting with four people from four different time zones, it's a nightmare. And when people are completely distributed, that can get to be a problem. So, one is some basic things like pegging to a timezone, really important. But the second thing is async video. Super important for certain aspects.

So, when people ask me for a one-on-one meeting right now, I tell them, send me a Vidcast first. And I don't think that that's an area that people think about intuitively today. We need to make sure that people get there. So, I think getting more wander time, thinking more of a fusion of sync and async, those things are pretty important.

Chet (15:01):
It seems like you are ahead of your competitors on the async concept. Is that true? Do you think you are?

Jeetu (15:09):
Yeah, I mean, yes, we are. I mean it's not even my opinion, it's fact. Our competitors don't have an async tool. We actually have one that's actually part of the suite. And the thing with async is it's not just about messaging and chat, which we have and it's great, but video async can allow you to infuse tone that's missed in chat. And so, when you have video async and I can actually send someone... Every couple weeks, I will send my team a message. I have 12,000 people on my team. I'll send them a message saying, "Hey, what's top of mind for me?" I don't rehearse it, I just get on and I actually talk for a few minutes on what things have been top of mind, what have I learned in my travels or whatever. And then they all respond and react. And I will get a few thousand views on that thing. And it's a great way to communicate with people.

Chet (15:56):
If I could go down the rabbit hole here just for a minute, there's not a single person who, whether you work in the office, you work hybrid, whether you work distributed, that wouldn't tell you they don't have meeting fatigue. What's your one piece of advice to groups on how to get over it?

Jeetu (16:15):
Let me tell you why meeting fatigue happens. Meeting fatigue happens because you are sitting in the same video angle, looking at the same people for an hour, hour and a half, two hours. It gets cognitively exhausting for the brain to just then have all the messages and the receptivity of the messages when you see the same visual angle. So, think about a two-hour meeting. In a two-hour meeting, you feel exhausted, but in a two-hour movie on Netflix, you don't feel as exhausted. This weekend I went to see the movie, Air. You got to see it if you haven't seen it, but one of the best movies. At the end of two hours, I was energized. I'm like, oh, my God, this was such...

Why is that? And so, we thought long and hard about this problem at WebEx and we launched something called cinematic meetings just recently. The reason you don't feel as fatigued in a movie as you do in real life, besides the fact that they're professional actors, they know how to engage you and all of that, is every five to seven seconds the camera angle changes. And so, what it does is cognitively your brain resets and then you have a refresher on your brain every 5, 7, 10 seconds. Whereas on video, I see the same angle. And as good looking as you are, Chet, if I saw you this way for two hours, I'd get tired, right?

Chet (17:31):
I'm hurt.

Jeetu (17:32):
I know you're hurt. So, what I think we need to do on this front is actually have this construct of cinematic meetings. So, what we've done is used AI in this case for video intelligence to be able to detect movements of people, merge all of the cameras to operate as one system. And if you turn around, you can actually now look at a different camera. That camera knows it's looking at you, it'll actually pick you up based on your speech and your video, and you can now have multiple angles. It's like you're watching a movie.

Chet (18:11):
Shifting gears, Cisco recently announced a partnership with Mercedes-Benz, which sounds just awesome. When do you think we really transformed the car into a home office?

Jeetu (18:21):
I think it's already starting to happen. So, if you think about what Ford did with the F-150 and what Mercedes is doing, what the other car manufacturers are doing. Here's what's happening in the auto industry, which is fascinating. Every auto manufacturer right now is thinking about EVs as the number one priority. That's not the news. The really interesting part is they aren't just thinking about the combustion engine being replaced with an electric vehicle. They're fundamentally re-imagining the auto experience as you're doing that. And so, it just happens to be this one catalytic thing of moving to EVs, but the screen real estate in the cars has changed. The way in which you can actually sit in the car and not feel guilty because your engine's running and actually use that as a home office, that's changed.

The car being used as a battery for your home to generate, give back, reverse power back to your home through the grid, those things are starting to happen with the Ford F-150. So, what you'll start to see is this kind of notion of technology being fused into this piece of hardware that they call the car, is actually getting to be very, very prominent. And it's being used for use cases way beyond take me from point A to point B. It's being used for use cases, which is I now have a separate room that I can use if I don't have a place that I can take a call in private. I now have ways that I can actually engage with people in a conversation in my car because my car is my mobile office. I can now have autonomous driving, and so I'm going to have that hour to go out and do things in my car that I didn't have before.

Those are pretty exciting innovations in the auto industry. And frankly, people ask me like, "Hey, what are the cool industries that you'd want me to think about?" Of course, one of them I'd say is hybrid work. One of them I'd say is security. Because I run those businesses, those seem very interesting to me. One of them is AI, but the other one is auto. I love cars, but never has there been a time when the role of the car is going to change so much. And it's beyond just car being an identity of how cool you are, it's actually the use case for the car is changing.

Chet (20:35):
I would agree. And it just feels like it's still early days and the reason is because it's not being reinvented.

Jeetu (20:40):
When there's a market shift that happened that can be classified as a megatrend, you as an incumbent have one of two things that you can do. You can either fight the megatrend or you can use it as a tailwind. Everyone will collectively innovate more. I'm actually quite excited to see what the auto industry has in store for us over the course of the next five, 10 years. It's pretty exciting to see what they're coming out with.

Chet (21:02):
That's awesome. All right. Rapid fire questions brought to you by GPT, obviously. If you could trade places with anyone for a day, who would it be and why?

Jeetu (21:15):
I'd like to see some of the world leaders that believed in technology, but many, many years ago, like Kennedy, with even what Abe Lincoln was doing with inclusivity and ending slavery and all of that. If I was a living person, I'd say the obvious ones that I'd really like to get more in the heads of are our folks who are building the next gen. It's very cool to see what Sam's doing at OpenAI.

Chet (21:42):
Describe your leadership style in one word or phrase.

Jeetu (21:46):
Direct.

Chet (21:47):
The most important part of your morning routine.

Jeetu (21:50):
Saying good morning to my daughter.

Chet (21:52):
Last thing that made you laugh out loud.

Jeetu (21:55):
I think the movie Air had some funny moments in it. I just saw it this weekend, that was pretty funny.

Chet (22:00):
Best piece of advice you have ever given.

Jeetu (22:04):
The quality of the problem that you picked to solve is directly proportionate to the success of the outcome. Harder the problem, the higher the likelihood of success.

Chet (22:14):
Best piece of advice you ever received.

Jeetu (22:17):
You can learn anything that you put your mind to.

Chet (22:20):
And you practice it. Jeetu, this has been phenomenal. It is a pleasure and an honor to have you on the podcast. We really appreciate your time.

Jeetu (22:29):
Thank you for having me. It's always good to see you, Chet. And congrats on all the great stuff you're doing in your business and growing that. I'm really happy to see people like you continue to keep succeeding.

Chet (22:37):
Thank you so much for tuning in to the Inspired Execution podcast. If you enjoyed today's episode, please like and subscribe. We have many more phenomenal guests and inspiring stories to come, so be sure to join us next time.

Inspired Execution

Jeetu Patel

EVP & GM, Security & Collaboration, Cisco

Asynchronous Work in the Age of AI with Cisco's Jeetu Patel

Episode Transcript