Media Week NYC theCUBE + NYSE Wired | Navrina Singh, Credo AI

Clips
More from Media Week NYC theCUBE + NYSE Wired

Navrina Singh

Founder & CEO

Credo AI

play_circle_outline Enhancing Trust and Compliance: Credo AI's Governance Platform for AI Oversight and Risk Management

play_circle_outline Promoting Collaboration Across Expertise for Addressing Emerging Risks in AI Systems

play_circle_outline AI's potential to make humans more productive, but only through responsible AI

Info
Transcript

Navrina Singh, Credo AI

Navrina Singh

Founder & CEO Credo AI

TheCUBE's coverage in New York City continues at the New York Stock Exchange. Navrina Singh, founder and CEO of Credo AI, discusses their AI governance platform and its importance in today's AI landscape. The platform helps with oversight, risk management, and compliance with AI regulations. Singh's background in hardware engineering and experience with AI safety led to the creation of Credo AI. The platform targets chief data officers and chief information officers in organizations to help them adopt new AI capabilities responsibly. Singh emphasizes the need... Read more

explore Keep Exploring

What does Credo AI, the world's leading AI governance platform, help with? add

What are some challenges with artificial intelligence and why is it considered a multi-stakeholder initiative? add

What is the importance of responsible artificial intelligence in ensuring that humans can benefit from increased productivity and free time? add

bolt Powered by CUBE AI

Navrina Singh, Credo AI

search

>> Welcome back, everyone, to theCUBE's coverage here in our New York City studio on the floor of the New York Stock Exchange. Actually on the balcony looking out over the floor. Great location. The market's behind us. I'm John Furrier, host of theCUBE. We're excited by bringing this New York access point. TheCUBE East, we're calling it. That's where all the network action is, local to New York. But there's tons of events happening. Of course Silicon Valley and Boston part of it. We've got a great guest here who was also giving a tech talk upstairs. Part of the NYSE Wired community collaboration, tech talks here. Obviously CUBE interviews and ongoing coverage. Navrina Singh, Credo's AI founder and CEO is here with me. Great to see you. Thanks for coming on.

Navrina Singh

>> Thank you so much for having me, John.

>> So you just gave a tech talk upstairs in the 1792 private room with a bunch of thought leaders and experts and potential customers, but really it was a community event. Love the engagement there. Your presentation was in probably what is becoming the hottest area that we're seeing on SiliconANGLE in our research coverage, is if you asked me four years ago if governance was a big deal I would've said, "Of course it is, in that small little vertical." But now, with AI, everyone's talking about it. You guys are doing something pretty compelling in risk management, governance. And really the key to AI is actually having the scale to make data work. Sounds easy on paper, but there's a lot of work to go in. So explain what Credo does. And we'll get into some of the cool things you presented.

Navrina Singh

>> Yeah, absolutely. Credo AI is world's leading AI governance platform. And what that means is we help you with three things. One is we help you with oversight of all the new AI capabilities, like generative AI. Secondly, we help you with risk understanding and management. And third is we help you get to compliance with the plethora of AI regulations and standards that are showing up in the market. We started the company about four and a half years ago, coined the category AI governance, and for the past four and a half years we've been at the forefront of not only delivering a standardized software that helps companies with this oversight and accountability, but also helping companies use governance as a competitive advantage to win more business, to retain customers longer, and to really build that trust in this very dynamic AI ecosystem.

>> Yeah. I want to get into the whole speed advantage of how real-time insights and real-time value. Seems to where AI is accelerating. Everything's about accelerated computing, accelerated value. Nvidia talks about that. I think they're right on categorically. But first, what was the origination story? How did it all start? Take us through where you guys are at. Funding, customers. But give us the origination story and an update.

Navrina Singh

>> Yeah. John, Credo AI is not a idea that one day I woke up and said, "Let's do it." It is actually the outcome of multiple different things that have happened over the past 20 years in my career. I'm a hardware engineer by training. Spent early years at a company Qualcomm building cell phone chips, and then moved into product management and then into robotics. And this was way back, like 13 years ago, when ImageNet came to being. And we were really trying to think about, how do you build computer vision models on a small form factor but very powerful form factor like mobile compute and connectivity? And I think that was a little bit of an aha moment because when you have these robotics applications working alongside humans in manufacturing plants, physical safety. You have these collaborative robots working alongside human. Physical safety becomes a pretty big thing. So 13 years ago, started to really think about, what does AI safety mean? And what is the evolution, especially as a lot of these applications become more I would say powered by machine learning application? Fast-forward, was hired into Microsoft to think about cognitive services, which were a set of services from facial recognition to speech to NLP. And what does commercialization of those look like? And as you can imagine, the scale and speed of AI has not slowed down at all.

>> Yeah.

Navrina Singh

>> And so we started to see the challenges if you didn't have the oversight and the compliance requirements early on in your ML ops pipeline. So the past 13 years, it started out with the physical safety of humans, then moved into digital safety as these very massively powerful machine learning systems were coming to being.

>> Yeah.

Navrina Singh

>> And so the key question was, is there a software solution that can actually help companies not only be accountable but make sure that there is standardized oversight across these very powerful systems? And so Credo AI was born from this massive pain point that I saw in the market. And once you see a problem as a founder, you can't unsee it.

>> Yeah, yeah.

Navrina Singh

>> So I was like, "Let's go. Let's bring not only the software horsepower but the AI horsepower to make sure these systems are safe and responsible."

>> Yeah. It's interesting, what are blockers today. Once they're removed, like I said, things accelerate. So let's get into the product and what you guys are offering. What's the core problem that you're solving? Unblocking the grinding out of all the procedures? Does it actually bring a service model to, say, CFOs? Who's your target persona? Take us through, because this is an area that categorically is growing in relevance.

Navrina Singh

>> Yeah.

>> And it's different by company.

Navrina Singh

>> Absolutely.

>> How would you talk about who your persona is? Who do you sell to?

Navrina Singh

>> Yeah.

>> Who's the value user?

Navrina Singh

>> Yeah. John, the best way to think about Credo AI's software is we help organizations first and foremost adopt new AI capabilities fast, with eyes wide open, because they can understand the risk of these powerful systems very quickly. And then the second is we obviously help them get to compliance where needed. So the key buyer for us is a chief data officer or a chief information officer within an organization who is responsible for your AI footprint and AI center of excellence.

>> Yeah.

Navrina Singh

>> And as you can imagine, our platform really is trying to bridge the oversight deficit that exists between your AI ops, your LLM ops pipeline, and your business pipelines, the traditional governance risk and compliance.

>> Yeah.

Navrina Singh

>> So our buyers in the past four and a half years have been very focused on making sure that... the problem for them is, I know that AI can get me all these ROI and bring me all these top line growth, but how do I do so in a responsible way by building trust, by understanding risk, and being compliant along the way?

>> What's interesting is that the CDO, chief data officer, is usually that person who's been handling a lot of the data science, analytics, the data engineering. Now, with data and generative AI opening up more surface attack vectors on the surface area, you're seeing security.

Navrina Singh

>> Yeah.

>> Because security is bundled up into a data and risk management problem. So security now is involved, so sometimes CISOs. So this is now not a department.

Navrina Singh

>> Yes.

>> So I want to get your thoughts on how you see that evolving, because there's CDOs with almost a decade in of, I would say, mature learning and practices. How does that flex now with gen AI? Because now the stakeholder is getting horizontally across the organization. Why? Because the observation space for data is expanding. So it's not the analytics department anymore. Oh yeah, I got the pipeline. Cool. Check. But now it's programmable.

Navrina Singh

>> Exactly.

>> So those robots we were once worried about is just all software, and that could impact the business model.

Navrina Singh

>> Oh, absolutely. John, I think you've hit the nail on the head. One of the challenges with artificial intelligence which, by the way, even the most advanced AI company is going through, it's a very multi-stakeholder initiative because there are so many different kind of challenges within these AI system, from security to privacy to really thinking about socio-technical impacts. So one of the things that we have found in Credo AI is, even though the buyer can be a chief data officer or a CISO or a CIO, our sales cycles are very multi-stakeholder. So you do need the risk and compliance perspective. You need the data science perspective. And I think that also the exciting part of that is that just shows that the surface area not only has increased but, because of the power this technology is going to bring to a business, we still have to unlock a lot of ROI from it. But the power of this technology is so massive that you do need all these different expertise coming in to solve big challenges.

>> I saw Pamela Gupta, who's from Connecticut. She was in your session upstairs. I interviewed her. I like her line because she's like, "I'm a consultancy. I teach you how to fish. I don't catch the fish for them." I get that her role. But she's smart and she brought up a good point that I didn't think of and made me connect a few dots on the fly, real time, was that some of these new questions bring up new risks.

Navrina Singh

>> Oh, absolutely.

>> So now you've got the models with the LLM, whether it's computer vision or other. The fact that there's no answer yet is people almost don't want to know. I don't want to know. Don't even say the word.

Navrina Singh

>> Yeah.

>> They look the other way. They're trying to figure it out. It's so new.

Navrina Singh

>> Yeah.

>> So AI is opening up more risks. So when you go to a risk compliance department, when they hear risk, they have to go down. That door is open. They got to go to completion. So you guys built a platform that does that, helps that. What's changed that you guys can bring to the table? Because I see a lot of pain on, how do I get my arms around it? To, okay, what do I do and who's going to own it?

Navrina Singh

>> Yeah.

>> And what's coming up a lot is this idea of not accountability but raising your hand and saying, "I own this." Someone has to own it, and so at some point someone has to run your platform.

Navrina Singh

>> Yeah.

>> Who? Who?

Navrina Singh

>> Yeah. Right now it depends upon the AI maturity of the business. Most of the organizations, I would say it is the chief data officer or chief AI officer who runs our platform. They are the budget authority. But again, John, very important to understand, because of the socio-technical nature, you literally do need to have data science and you need to have risk compliance policy actually using this platform. And this is where our integrations play a massive role, because we are integrated into all your AI infrastructure and business infrastructure so that we are making it easy for all these multiple stakeholders to just work together. I do want to address something that you were I think very well put together. One of the things with governance right now is these emergent risks.

>> Yeah.

Navrina Singh

>> And as AI experts, one of the things we do at Credo AI, and that's one of the reasons why we are the leaders, is we work very closely with standard-setting bodies, with institutions like Stanford High and others to make sure that not only what's happening in the regulatory ecosystem we can codify within our platform but us, as AI experts, we are red-teaming these systems, testing these systems, trying to figure out new emerging things that we can very quickly codify so our risk and compliance stakeholder does not have to.

>> Yeah, yeah. I love what you just said there, red-teaming. And that's a security concept. But red-teaming is important because you're not doing it for just security testing. You're doing it to understand the nuances of how the game is being played. And Anand from ICE here, who you know, runs the technology? He and I were talking. And this is the theme. And I want to get your reactions because I think people are seeing this now, but I want to over amplify it because you can't play a game you don't understand.

Navrina Singh

>> Yeah.

>> So it's like playing chess and not knowing what checkmate is. So you got to play the game. So people need to get their hands dirty. And what's interesting is that you're doing two things very hard. You're bringing technology stack to the table with a solution, but the cultural process mismatch has to be addressed. That means the front end and back end theaters are evolving very rapidly. I've never seen this in my career. I've seen many ways of innovation. It's either a front end or a back end innovation. Cloud was back end, web was front end, mobile was front end, cloud was back end. Now both environments, which means it's harder to get it right because you got to do process culture and full tech stack, managing data availability and new configurations, how resources are allocated in the three layers of the stack.

Navrina Singh

>> Yeah.

>> Do you agree? Because you have to do it. If you don't do it-

Navrina Singh

>> You have to.

>> You can't.

Navrina Singh

>> I love that framing, John. And that's a framing actually we've used since the beginning of Credo AI. We always talk about, how can software solve three layers of trust problem? The first layer of trust is with the ecosystem. Do you understand the regulations? Do you actually inform those regulations and standards? And can you bring that in? The second layer of trust, absolutely spot on, is the people and processes because, within an organization, unfortunately the traditional way your software development lifecycle now needs to completely evolve for AI ops, as an example. So what is the not only cultural aspect but also the literacy aspect that you need to infuse into your people and process? Very important. And then the third is the tech stack, right?

>> Yeah.

Navrina Singh

>> So this is where, Credo AI, we are very focused on the AI use case level governance because, especially when you think about these foundation models which are very dual purpose, they can apply to different use cases. And so you have to think about, okay, what's the impact? But then also trace it back to the models that's powering it, the data sets that's powering it, the people behind the problem, how it is structured. What does the business care about? So there's these layers of trust that you have to start building.

>> Tell me what you're most excited about right now, because you've got a great opportunity. It's an area that's been evolving. It's been your career. You saw the problem. Because I love that line, "Once you see a problem you can't ." That's so true.

Navrina Singh

>> Yes, exactly.

>> It's like stuck in your head. You got to solve that. And it's a big market, and it's hot right now. And it's super relevant technically for how data is structured. And so all this is built in at the beginning, not bolted on after. So I love your mission, but what are you most excited about? What energizes you?

Navrina Singh

>> I think artificial intelligence and how it is going to help humans become more productive is what energizes me. But I do want to emphasize, the only way we can get to that outcome is if we are using responsible AI. So this is where Credo AI's role in this ecosystem, how do we become the seal of good artificial intelligence? Is so critical because all the benefits, the upsides that we are going to see with AI, whether it is coding gains, whether it is productivity, that I have more free time. All that is going to be possible only if we can trust this technology.

>> Yeah.

Navrina Singh

>> And the trust actually just doesn't happen. It has to be earned. And so what we are seeing is right now the mission that we are on to become the seal of good AI governance. It is one of the most critical things to build that infrastructure of trust for AI.

>> Yeah. You guys have got a great tech stack. Love the solution. Give us the business update. How is it going?

Navrina Singh

>> Yeah.

>> You guys, you feel good about where you're at revenue-wise, customer, funding? Give us an update on the business and business model.

Navrina Singh

>> Yeah. We started the company in 2020, so four and a half years in. We are growing very rapidly as a team. We have 3Xd this year. One of the core things that we are excited about is the kind of investors we've been attracting to our business. We've raised about $42 million so far with investors like Sands Capital, Decibel, AI Fund, Mozilla, FPV, and others who are really not only excited about the potential of artificial intelligence but understand the power of responsible AI. We serve global 2,000 companies in multiple sectors. Financial services, healthcare, pharma, government. Some of our customers include MasterCard, Northrop Grumman, Booz Allen. Our partners include McKinsey and Databricks. So as you can imagine, right now there is a need in this AI ecosystem for a good governance, a trusted governance partner. And Credo AI is that for the ecosystem.

>> Thanks so much for coming on theCUBE here. What do you think about our NYSE studio? CUBE East, this access point to the network?

Navrina Singh

>> I'm just so excited because this is where magic is going to happen. And giving us, startups like Credo AI, a chance to come and share our story. Hopefully one day we are going to be listed here.

>> Well, I'll see you back in the Bay Area. I'm heading back tonight.

Navrina Singh

>> Yes.

>> It's been a great media week. Thanks for coming on.

Navrina Singh

>> Yeah. Thank you so much, John.

>> All right. I'm John Furrier of theCUBE. We're bringing it down, all the action, to you from the New York Stock Exchange floor in collaboration with NYSE, Brian Baumann, and the Wired community. Again, this is going to be our East Coast hub, our access point, point of presence, super node for all the smart people here in New York connecting to Silicon Valley. I'm John Furrier, your host. Thanks for watching.