From Cloud Pioneer to Al Leader: Salesforce's Journey to Agentic Al with John Taschek

How are intelligent platforms fundamentally transforming the way we work, right now and into the future? 🔮

‍

John Taschek, Chief Market Strategy Officer at Salesforce joins host Daniel Newman as a speaker for the Enterprise AI track at The Six Five Summit: AI Unleashed 2025. In this session, they dive into Salesforce’s progression into enterprise AI, along with the latest trends, groundbreaking innovations, and the future of work in the age of Artificial Intelligence.

‍

Key takeaways include:

‍

🔹The Rapid Evolution of Enterprise AI: Explore the incredibly fast-moving landscape of enterprise AI and its profound impact on core business strategies, moving beyond simple automation.

‍

🔹Salesforce's AI-Driven Innovation: Discover how Salesforce is strategically leveraging AI, including the progression to Agentic AI, to dramatically enhance customer experiences and drive operational efficiency across its platforms.

‍

🔹Building Trust and Driving Data-Driven Decisions with AI: Understand the paramount importance of ethical considerations and building trust in AI applications, along with how AI is enabling smarter, data-driven decision-making within organizations.

‍

🔹Untapped Opportunities & Future Outlook for Enterprise AI: John shares his insights into the vast, yet-to-be-fully-realized opportunities for AI in the enterprise, including practical advice for leaders transitioning from experimentation to full-scale AI implementation.

‍

Learn more at Salesforce.

‍

Watch the video below at Six Five Media, and be sure to subscribe to our YouTube channel, so you never miss an episode.

‍

Or listen to the audio here:

‍

‍

‍

Daniel Newman: Hey everyone and Welcome to The Six Five Summit: AI Unleashed. I’m joined today by John Taschek, Chief Marketing Strategy Officer at Salesforce for an Enterprise AI spotlight on how intelligent platforms are shaping the way we work. John Tashek, welcome back to The Six Five Summit. So great to have you. I always love the chance to sit down with you.

‍

John Taschek: Thank you, Daniel. It's. We have the best conversations ever, so I'm glad to be here.

‍

Daniel Newman: First of all, you know, John, you've got a big remit, you know, leading market strategy and more. You've been around, by the way. How. Just for everyone out there. How long have you been at Salesforce again? Just, Just give me that number.

John Taschek: I don't want to give you the exact number, but it'll be 22 years in July.

‍

Daniel Newman: Okay. You have one of those little hour tickers.

‍

John Taschek: I do. It's a countdown clock. No, no, I love it here. So it's, you know, 22 years. It seems like the beginning to me.

‍

Daniel Newman: Well, you've been there for a minute and you've probably been through many iterations, instantiations, updates, changes, transformations, acquisitions, expansions, contractions, market pivots, shifts, geopolitical ups and downs. You've been through quite a bit. But have you been through anything that's been as fast moving, transformational, disruptive as this sort of AI agentic moment?

‍

John Taschek: No, And I thought it was, you know, when I joined the company, there were only 300 people at the company, and I thought that was like super fast moving to, you know, move from on premises into the cloud, on demand, or SaaS, whatever it was called back then. But into the cloud. No, this is much faster than that. It's much faster than the Internet, which I cover. It's much faster than database technology. It's faster than anything I've ever experienced. At the same time, the company is kind of more or less the same. You know, we're, we're addressing the market the same. We have the same values. We treat ourselves as a startup. I mean, there's well over 70,000 people now, but the culture is very similar. So we're tackling it, but it is coming much, much faster.

‍

Daniel Newman: It's interesting, probably a case study in itself of how you go from 300 to 70,000 and keep your culture intact, but I can say, and I haven't been around it for 22 years actually, John. If I go back that long, I'm still in college. I don't know if that makes me older.

‍

John Taschek: Not to date me. Not to date me. But yes, a lot's changed in 22 years for sure.

‍

Daniel Newman: But I will say there are definitely some aspects, as I experienced working with and interacting with Salesforce that can be very startup and some of the kind of culture of the ohana, the family, the sort of belief. You experience it every time at least I do every time I go to Dreamforce and sort of see how the company interacts and you know, we're sort of professional event attenders, John. You know, in my world, I probably make it to 70, 80 a year at least. And you do get to see a lot of different cultures. But you're also a company that kind of catches things first. There's this kind of line I still like from the movie The Big Short, where the character that plays Michael Burry, one of the great short sellers that caught that moment, goes, I may be early, but I'm not wrong. And then the guy responded to him, he said, it's the same thing. But Salesforce has been early on many things. Right on many things. It pivoted some things. For instance, it saw the opportunity of AI, enterprise AI very early and now it's had these different instantiations of it at Einstein early on and Genie. And now you have Agent Force, which is kind of like this progression that went on. But talk a little bit about the AI journey that Salesforce has been through for yourselves, for your customers. Like this has been a pretty massive pivot, but it hasn't been as fast for you as some companies you've been doing it for. I mean, it's been several years that AI has been really front and center at Salesforce.

‍

John Taschek: Yes, and you're right, and we were early. It's not that AI didn't exist in a predictive or rules based way before Salesforce, but integrating it into enterprise apps. Over a decade ago we acquired a company called Metamind. We had done our own predictive research, of course, but when we acquired MetaMind, we started a research lab. We did a lot of work and research on AI, especially predictive. But we also started getting into data architectures, LLMs, and prompt engineering. We have a lot of patents on that. That started years before Chat GPT was released to the wild, which is now just what, two and a half years ago roughly. That time is very compressed for me. But yes, the research lab has done an incredible amount of work in AI and integrating it into the enterprise applications to make it seamless to the customers who are using it and their customers who are using their solutions. And it's given us a huge competitive advantage. In a way, it doesn't even seem like a pivot, of course, from, of course, from the outside world. That's exactly what it seems like internally. It seems like a progression that's rapidly evolving. And because we had done this work in LLMs and prompt engineering, we were doing all kinds of things. We were doing gene research out of the research lab and published in Nature magazine several years ago. And if you call it a pivot, which I won't, it would be a very quick evolution that started actually four years ago when we were getting our data strategies in order, in order to make AI make more sense, and more usable and actionable and embedding that in all the, you know, the platform that we have, the platform and the clouds, you know, sales, cloud service, marketing, platform commerce, all the things that were traditionally called CRM. And when Agentic came out, I said, well, you know, the first instances that came out were co-pilots, which more or less added individual productivity, but didn't consider a lot with business policy and the way businesses work. So a very good leapfrog there for co-pilots. But Agentic gave us a whole new era that we were perfectly well set up for.

‍

Daniel Newman: If I can capture in a few, in a phrase, John, it's a bit about kind of this pervasive iteration is sort of why it didn't feel like a, you know, a pivot in so many ways, because it was sort of the layering on. It's very much what SaaS has always been, right, is the ability to sort of, you know, perpetual, pervasively update and change things to add what is needed. And as AI has come to the forefront, you can do that in SaaS because you don't need to rip and replace. It's a feature. It's the, hey, the next feature was being able to use a copilot. The next feature was able to use some type of ML to better maybe understand churn risk or deal likelihood of deals, closing likelihood. But it keeps getting more intelligent and the amount of data that it's able to use keeps becoming larger. And as that data becomes larger, then it's about picking the right data that's going to likely get you to the right outcome that's going to enable an agent to maybe perform the right task to do it concurrently. But you know, if there's one thing about Salesforce that you know, someone you know in your role and among your team really probably has to answer for in the future is going to be how does software change in this era? And agent force is a clear intent. You could say it's not a pivot. I would say it's a pretty big transformation of the company though. You're going to get more horizontal. You have to. Right. Salesforce has to say look, we don't have every application but we want our agents to be able to talk to every application. How are you sort of thinking about doing that so you can become one of the most, continue to be one of the most critical platforms that companies use day in and day out to run and make decisions in their business both agentically and with humans.

‍

John Taschek: Well, you're kind of asking an existential question and well, as well as a pragmatic question at the same time. When Salesforce started the business model was one of the innovations that we had along with the philanthropic model as long as delivery models, update model being a platform model, being agnostic through API model and continuous delivery roughly three times a year for the product for 25 years. And that would be pragmatic. We still do that except it's more often than once a month now changing from a model where it's per user per month and entirely predictable for you're paying for a seat and then you get access to the technology into a business model that involves more flexible pricing and consumption, that consumption based pricing. So that's one of the things that we're doing as far as the technology goes and the innovation goes, we obviously are still working on every single part of the business. Whether it's better sales or better service or customer service or better commerce marketing across the board, we're still doing that. Customers still ask for it, still want it. When agentic comes in, it's giving actions and autonomy to the agents so that they can work on your behalf. And in order for that to happen we had to do a bunch of foundational technology including trust. So if you look at the way that Salesforce announced the product lines we first came out with, well you said genie but that was a short lived name. It's a data cloud and the data architectures. The second thing we did. And the first in the agentic world was trust, a whole trust layer which did toxicity detection. In order for a platform to be usable and valid within a business and with business policy, it has to be trusted. Otherwise they'll use it and get rid of it. So the trust layer was incredibly important. And that was two years ago. And since then the agentic Agent Force which was just launched at Dreamforce, how many months ago is that? Eight months ago or something like that?

‍

Daniel Newman: Just yesterday, John.

‍

John Taschek: It seems like yesterday again. Time is very elusive to me these days. But when we launched Agent Force, which is the agentic part of Salesforce, the foundational parts were already, being worked on. You know, they're never going to be complete, but they were already already being worked on and delivered. So agentic makes more sense when it's, you know, combined with the rest of the architecture. And when you look at the market, because you mentioned like we have to go horizontal. Well, I firmly believe CRM is very horizontal. It's anything that touches the customer. Now the customer may in the future involve an agent, but a customer. And it's a pretty wide scope with a very large total addressable market. But we can go even farther than that. And if I could show you right now visually and visualize our Agent Force architecture, you won't see, you will see a very horizontal platform and it's already being used in that capacity.

Daniel Newman: I could talk to you about this all day, but the beauty of The Six Five Summit is these are sort of like rapid fire conversations, John. So I got one more for you. We assess very closely the agent space and we also did some interesting work with Salesforce on agentic AI. We found that there were trillions of dollars of efficiency. There's even more trillions of dollars in productivity to be gained. There's tons and tons of questions about what the future of work looks like, which that's a whole nother topic. But just from a standpoint of customers, since there are thousands of enterprise customers, watch the summit and many of them are the customers of Salesforce or enterprise software consumers. They're trying to deploy this stuff. I know at Dreamforce last year Mark was, he got up and talked about some customers that were already deploying agents getting value. What are you seeing now that you're eight months later, the product's further developed, you've built out more data, more trust. I'm really glad you brought up trust, John, because I think that layer sometimes is forgotten because of how fast we're moving but we really need to make sure we're not spilling data and doing this. But what are you seeing in terms of POCs turning to broader deployments, customers getting value quickly. How is that accelerating in terms of the, against the objectives that you have at Salesforce?

‍

John Taschek: It's accelerating so fast and it's changing daily. The customers seem to be aimed at efficiency at the beginning. Like how can I be, how can my workforce be more efficient? Sometimes that translates into can I let some people go and have the same capacity, same operations and there's, there's customers still out there that demand efficiency because you know they want it and they need it. They have to manage their own budgets and their own growth. What's changed is much more of a growth mindset is how Agentic can make the company scale broader using the same people or maybe skills developing, you know, some employees to have different skills and more strategic and it's much more growth focused and it's also going into industries that are fairly broad. Dreamforce we said hey, here's a book publisher. They needed to have this very streaky life cycle in their year. How do they do that, how do they just manage that and be great at it? Well that's one area we're seeing that when Agentic comes up and the way we're positioning it is that customers are no longer looking at a particular silo necessarily. They're not looking at just efficient customer service. Now we're our own use case in, in our own customer service. We have a huge ROI for ourselves in using Agent Force internally. And that's great, but it also is better. And your report and your methodology was incredible. Thank you Daniel for that. Showing that this is a multi trillion dollar market as far as the workforce is concerned and the total addressable market is just increasing when people start thinking out of their silos. And so that's the change and only in the last few months. Again, Agent Force is only a few months old, but we're seeing that very dramatically. And the total addressable market for agents I think you identified as about 5 billion a few months ago. That was, it's probably grown beyond that. 

‍

Daniel Newman: It's an opportunity that will probably land in the trillions just because of the cost takeouts and then the acceleration. John. I mean, look, our perspective is there are a lot of sorts of questions about how these things coexist with work and how we augment the workforce and how we then accelerate. But every, every industrial revolution has created more, not less. And so sometimes as we get into those sorts of, oh my gosh, this thing's going to write my research for me. We realize that we automated things like social media posting and it didn't get rid of the need for content creators on social media. Like there are things that happen. But AI is incredible. The pace that's moving. John, to your point. But one of the things I really want to do just before we sign off here is just it's really great to hear that customers and enterprises are getting value. We can spend a lot of time focusing on these sort of consumer use cases search, you know, LLMs and AI. But the ability for companies to deliver service more quickly or to be able to fulfill orders and products more efficiently or you know, to speed up drug discovery to, you know, engineer better designed cities, things that AI and agents will help is just very, very exciting. And of course in the end we all have to take care of our customers. So the best tools that enable us to be very responsive and supportive of customers building those long term relationships. And it's great to hear, John, that Salesforce is doing the work. It's got a deployed product in the field that it doesn't take years to get this out there and in fact in weeks and months that companies can employ and deploy this kind of technology. John, I gotta leave it here. We should definitely reconnect soon, talk more about this. I can't imagine a year from now at our summit where this will be at. Thanks so much for joining us at The 2025 Six Five Summit.

‍

John Taschek: Daniel said that so perfectly. Thank you for allowing me to be here. It's a true privilege.

‍

Daniel Newman: Thanks for joining us for this Enterprise AI spotlight at The Six Five Summit. Stay connected with us on socials and explore more conversations at SixFiveMedia.com/summit. More insights coming up next.

‍

Disclaimer: The Six Five Summit is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded, and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors, and we ask that you do not treat us as such.

‍