Data Security In Sales Messaging With Ambuj Kumar
Much of this show has been about using AI in sales in terms of increasing touchpoints, have more calls, and connect with more people on social media. This time, Chad Burmeister shifts the conversation into sales messaging, specifically its security component. Fortanix is one of a number of companies that offer services in this space. The company’s founder and CEO, Ambuj Kumar joins in on the conversation to explain what they do on a daily basis. Ambuj talks about the need for encryptions and other steps to protect the privacy and security of business ventures worldwide. He also talks about how Fortanix is uniquely positioned to serve that space. Tune in to learn more.
Listen to the podcast here:
Data Security In Sales Messaging With Ambuj Kumar
I’ve got a guest where we’re going to talk about something that’s out of the norm of what I’ve been talking about. A lot of times, we talk about AI for sales and different technologies that can help you increase your touchpoints, have more calls, connect with more people on social. We’re going to shift a little bit and talk about the importance of security across sales messaging and how companies like Ambuj’s company are helping companies do that. It’s called Fortanix. Ambuj, welcome to the show.
Thanks, Chad. I’m looking forward.
Before we dig into technology and security role in and around AI, and how AI plays with that, I’d like to help connect the audience to you as a founder. I like to ask this question about when you were younger, some of your first memories, what were you passionate about? Did you always know you wanted to be a CEO? What did you do when you were younger that you enjoyed and were passionate about?
I was a little nerdy but I was in Math. I remember I would complete my year’s work of Math in the 1st and 2nd month, then I'll try to get access to next year's books or read magazines. I was not very hands-on. I didn't do many projects, but I was always intrigued by abstract ideas, the mathematical nature of numbers and things like that.
AI is the most powerful force we have in building and controlling data.
You would like this new game that I’m playing. My son is in college at the School of Mines and he’s going to be a Computer Science Engineer and he’s extremely mathematically oriented. The game is called Splendor. When you play the first batch, it’s a math equation. He’s like, “Dad, you can either take three chips or buy a card or you can reserve a card.” Reserving, you only get one chip. He’s like, ”Dad, you’ve lost two points in the overall scheme of things.” It’s like card counting in Vegas.
You then start to add these ad-ons and then you can reserve a card or you can block other people. There are all kinds of advanced formulas. Life isn’t as simple as the first version of Splendor. It’s got expansion packs. When you can have that level of understanding of the Math, it changes your ability to deploy products and think about things in a very different way. I’m sure as kids, a lot of times, we don’t necessarily think of how it’s going to help us in the future but it most certainly does. How does it tie? If you think of your math skills then where you were able to get a year ahead of everybody else, how do you use those math skills now?
You can only connect the dots looking backwards. At the moment, you'll never know. How it worked out for me was a few years ago, I decided to start my own company in security. As you know, security is on top of the minds of everybody, from us to our parents to prime ministers, presidents and CEOs of global companies. The way Fortanix addresses security is by using encryption and encryption is taking some stuff you have and gobbling it up so that if you don't have the right keys, you cannot read it or access the information. Encryption is a mathematical process. There's deep math behind that. It came full circle for me.
Thinking of where AI plays in that. I don’t think converting it and putting a key on it would be classified as AI but I’m sure there are some interesting algorithms and other ways that AI is used. Talk to me about how it’s used in your product, and then let’s talk a little bit about, do you use AI in your sales motion at all to try to reach new customers?
In our generation, the most powerful and transformative technology is AI. It is going to impact every aspect of our lives. From our perspective, AI requires the collection of lots of data. You want to analyze your models. You want to train them with data and things like that. Whenever you think about collecting production data, people are worried about privacy and security. We come in and fit nicely there. I’ll give you one example. We are working with the University of California San Francisco research team. The UCSF research team is one of the most respected medical institutions in the country. They have five Nobel laureates and they were the first ones to build up AI that got FDA approval.
If you have your chest x-ray, their AI can read it and make a certain determination. You can imagine that is the life savior in the COVID era because lots of patients, especially in developing countries, don't have access to technology, etc. Having software do this base level of determination is very helpful. UCSF team realized that making healthcare patients’ healthcare data available to AI was a very difficult task because of security and privacy. They partnered with us. Fortanix and UCSF are working together so that if you are an AI developer and you have some brilliant idea of finding a cure for cancer or what causes diabetes or whatnot, you can have access to all the patient data that you UCSF has without compromising on patient’s identity or privacy.
There are two things that come to mind there. One is on LinkedIn, a friend of mine posted the other day. He’s the former SVP of ZipRecruiter, Kevin Gaither. He posted, “Does anybody have real data about social selling?” There’s all this qualitative junk that’s like, “It doesn’t work anymore. It’s amazing. You should try it.” I went to our team and I said, “Let’s anonymize all of our data and let’s present the master roll up and then start to make heads or tails of what that data is.” What we found is that LinkedIn outreach across all of our customers in 2020 had a 29.16% reply rate. That’s huge compared to email at 2% to 3%. That’s a big finding in itself. We found things like when Michelle from the East Coast at SalesGlobe sent a note out to talk about sales compensation on a weekly basis to a bunch of executives, 72% of people replied.
Everybody has a challenge that they want to address and AI can help with that and make our lives better.
There’s a lot of information hidden in the data, but anonymizing it makes sense. One other thought that I had related to the data and being able to present it is taking the amount of time it takes to analyze the data. One of the solutions that ScaleX offers is what would normally take one year for a human to go out and click a million buttons, the technology can do in about 8 to 12 minutes. There’s no way that a company would hire that human to go do those million clicks. By collapsing a year of work into eight minutes, it starts to open up the expansion pack or that Splendor game like I talked about before. Your product without the anonymizing factor that companies like UCSF are comfortable with, none of that’s possible because you would be exposing PHI and data that you’re not allowed to expose.
AI is the most powerful force we have. We need to build up on it and the way to do that is through data. The number one thing that comes the moment you ask people to share their data is, “Would I lose control of that data?” If we can solve that, then we can connect the right buyer to the right seller. We can tailor our message so that it appeals to you. Let’s say that you are somebody with a degree in engineering. You are a decision-maker of my product and I want to pitch my product. If I knew the information that you’ve studied physics or you’re a biology major, I can use that information to tailor-made my email or LinkedIn outreach just for you. You cannot do that manually because the economy doesn’t work, but if I have an AI that learns and goes on LinkedIn and has all the information about you, I can use all that intelligence and create something that targets you uniquely.
We’re beta testing one of those players right now in emails for a customer. We pulled 6,000 data records. The append is running as we speak and we’re going to have up to twelve personalization sentences available to the salesperson, so that when they go to send, it’ll say, “Here’s the top two. Go ahead and pick from a list, which one you’d like to use.” It still requires human oversight because occasionally, the AI could say, “I saw your post. That’s cool. Congratulations,” and you just crashed your Tesla. That wouldn’t be a good post but the AI might not be smart enough to not congratulate you on that one. We talked a little bit about AI in sales and that is that powerful messaging mapping. Any other technologies you’ve seen that you are or your sellers are using to reach out to customers?
As a buyer, especially when you are making a significant personal decision, people want to research, “What are your options?” They will go on Google and search for related products. They will go on different websites to learn about it. They will go on LinkedIn to see what their friends are doing. Building a tool that collects all that information across the website, etc. allows sellers to know what is happening. That’s a great way of feeding the leads to our salespeople. We use that. Once you are pitching to a particular person, that person is also doing their job. They have their own individual context. The company they are working for is going on its own journey. Everybody has a challenge that they want to address. That’s why they are talking to you. Organizing all that information using AI so that it does not take five hours of my time to pitch for 30 minutes is of tremendous value.
There’s one product called Crayon. They look at websites like way back time machine, where you can go look at sites from the past. This tool will look at everything on a site and if the price changes. For example, if you’re a RingCentral and you’re competing against Zoom, it’ll tell you, “Zoom just dropped their price by $5,” and send you a notification. Not even for the sellers but for the VP of marketing and the head of a strategy can say, “They dropped by $5. Maybe we better do a buy twelve get a month free or drop the price by $5 and match it.” You’re right. It’s all of the ability of AI.
If you were to compare the internet to AI in terms of the market cap of companies. I think of when Bill Clinton spoke one time at a Dreamforce Conference. I remember he said, “When I was president, I grew productivity by 5% because of technology.” He said, “5% productivity related to lots and lots of money on the GDP and growth and everything.” I’ve always been wondering at what point does AI flip the switch where we’ll see a much bigger than a 5% productivity gain? Do you think it's bigger than the internet or the same based on your assessment?
Imagine that the person who first found a way of building swords using metal or the person who first discovered stainless steel because it has more strength. You go to a battle and you would slay your enemy because their swords won’t be as strong as yours. You are not going to share that information with everybody else, but the other side will see that there’s some magical material has come. They will also push their scientists to do something. What happens is if you have some powerful product or tool or technology, lots of disjointed efforts come in and people are trying to do their own thing.
If the productivity curves increase because of AI, the need for humans to deploy those becomes a little bit lower, thus helping the business owners to save.
In security, for example, I’m using AI to make my product better. UCSF is using AI to make its product better. In sales, people are doing the same. Right now, we are at a stage where people know how powerful AI could be and they are using it for their own niche, segment, vertical and company. Going back to your internet question or comparison, Cisco became so big because they were selling a router that appealed to everybody and every website operator, regardless of what their website did. I think the productivity boost from AI would far surpass what we got from the internet. We are at a stage where everybody’s trying to build their own website. They have not gotten to a point where all these websites are fine, but you need infrastructure, routers and DNS and all those things. Companies and entities that build that underlying infrastructure for AI we're creating massive companies but as a society, we are already benefiting.
That’s well put. I totally follow. It’s like the standards. When you come out with IEE and people say, “We’re going to standardize on that,” or HTTPS or all of the different internet protocols. When you can standardize AI around that platform-based, I think Watson and some of those have attempted to do that but not necessarily had universal adoption yet. To your point, that’s internet versus AI. AI should be an exponential gain because the internet’s just the plumbing. What sits on top of it is where all the magic happens.
Even now, if you ask some person whose life was saved using AI that UCSF has developed, they will have the internet for watching videos. This is real stuff. There are thousands of researchers finding a cure for cancer or even when you are driving, AI tells you whether to take left or right. Imagine you are rushing for some important event. It may not be like running to the hospital but even if you are late for your daughter’s game. AI does help us make our life better. It’s just that since it’s not standardized, we don’t realize how much value it’s adding but it will continue to grow.
I had a company that’s out of Arizona. The head of sales was on this show one year ago. She said that if we can plug into hospitals and get very little information like when did you last see the doctor? What were the 1 or 2 pieces of information that came from that? When you look across a sample of a company, let’s say 100,000 employees and you get a little bit of data, we can predict very close to at what point is that person going to get thyroid cancer or things that seemingly would be, “What do you mean? All you collect was two pieces of information in the last year on me as a person.”
When you pull it all together and aggregate it, imagine because of tools like yours, you get way more than two pieces of information about the person. You get it all. Putting that into data set at scale, lookout. Maybe that’s why Bitcoin was invented, to create super-fast graphics cards that can process this level of information. Maybe it’s all just a scam. What do you think of the future? If you think a few years out, how does the world look different or how do we behave differently in that new environment years from now?
I think all the things that can be automated will be automated. Humans will still do creative work. Higher-level understanding to your point, you can build the most powerful AI engine but it’s not you. Similarly, on the other side, it’s some humans and humans can understand humans much better. You need to direct that. I think big sales, security, transportation, healthcare and tourism. All those industries will fundamentally be transformed. They are already going through a transformation through AI and that will continue.
It seems to me that if you look at human interest stories like universal basic income, I’m going to bring it up even though a lot of people I know would be like, “What do you mean?” Conservative, liberal, all this stupid stuff. Let’s be real. If all of a sudden the productivity curves up here, the need for humans to deploy whatever it is we’re doing as people becomes a little bit lower like a lot lower over time. There’s a point in time where you can say, “Let’s make sure everybody on planet Earth is taken care of,” because AI has gotten so productive. I think there are a lot of smart people. I did a show in person in Colorado. We talked about that. She was talking about human interests kind of things. It seems to me that it’s not too far off. When you can have a computer do the work of ten humans that would normally take a year in ten minutes, pretty quickly, you can start to make some interesting decisions. Sometimes I think that’s the big chess game that’s being played. Unfortunately, it has to be played through politics.
You can also imagine a world where there's one giant robot that is solar-powered. It's taking all the power from the sun and it's powerful enough to do everything that is needed for you. Take care of all your physical needs, all my physical needs and everybody else. What would it do to our economy? Would we still be required to work or do we have to pay to work because we’re in a world where all are priced at $0? The person who gets to color the robot would have to please everybody. The situation might change.
Healthcare is one of the big places that work with you. What are some other industries? If someone’s reading and said, “I need that level of encryption,” what companies might they work for?
Financials are a big one like banks, insurance, security companies, IT services. You have your own web service where you are collecting customers’ information. You are manufacturing. Anybody who is affected by HIPAA, GDPR, CCPA and Schrems II, we would have something for them.
Do you have investors yet or are you self-funded?
We have Silicon Valley investors. Intel is one of our investors, Foundation Capital and Neotribe Ventures. We are well-funded and we have dozens of Fortune companies. It’s a big thing.
I appreciate your deep dive into security and anonymizing data. It makes people think and understand it at a simple level. Thank you, Ambuj, for sharing. I’m glad to have you on the show.
Thanks.
Important Links:
About Ambuj Kumar
Prior to founding Fortanix, Ambuj was the lead architect at Cryptography Research Inc. where he led and developed many of the company's security technologies that go into millions of devices every year.
Previously, he worked for NVIDIA where he designed the world's most advanced computer chips including the world's fastest memory controller.
He has a Bachelor of Technology from IIT Kanpur and a Master of Science from Stanford University, both in EE.