Fireside Chat: How Agentforce 2.0 Will Transform Selling

Our next guest is going to be Mark Benioff, the founder and CEO of salesforce.com. welcome, Mark.

Oh, it's great to be with you, Gerharden. It's always great to see you. How are you doing today?

I'm doing terrific. And I assume you in Hawaii right now?

Oh, I'm in my office.

Oh, in your office in. In San Francisco?

I'm in my office in Hawaii.

Oh, of course. And who is that gentleman over your left shoulder?

Oh, that's Astro, our agent.

And what kind of conversations do you have with Astro?

Oh, he knows a lot of things. And you'll find out one day. You know how exciting that is.

Gerhard, I want to zoom back to your very first programming job. You were creating an app that taught people how to juggle.

That's exactly right. Well, you have a very good memory, don't you? That was way back in 1979. Right.

And how long did it take you to create that app?

Oh, gee, I don't know. That was, you know, I. How long did it take me? I mean, that was way back in the day. I was 14 or 15 years old, and I was living in. In Hillsborough and Burlingame, California. And I was cleaning jewelry stores, actually, for Current Jewelry in Burlingame. And that was my job during high school. And then after my job, I'd walk across the street to the Radio Shack store and I'd program on the. What was the TRS 80 model 1.4k computer that had just come in. It was like $500. And I just thought that was the most exciting thing I had ever seen. Gerard. I'll tell you, it was really cool. And I mean, it was kind of manifest destiny, you know, that I could work on a computer and see it do something. But I just love that. But nobody really understood how it works. Certainly nobody in the store knew how it worked. There was no classes you could take. There was barely books you could even buy on software or how you know how to. How to write the software. So. And, you know, I was a young kid, so I saved up my money, and I ended up buying this computer and bringing it home. And you're right. I ended up writing my first software product, which was how to Juggle. And I sold it, which was amazing. Somehow I found a company to buy it, which was a company called C Load in Goleta, California. And they bought it for $75. And that was the first piece of software that I sold. Now, do you know what C Load means? Gear heard? No. Okay, so that C Load, that was.

Insert a cassette, right?

That was cassette load. So we would load the software with cassettes until we had a little cassette player there, you know, which is like a little tape player. And we'd be able to record and store the software. And they had a magazine like Selling Power, but it was on the silo, was on a cassette called C Load. And every month you'd get, you know, 10 or 15 pieces of software that.

They bought on their magazine.

You know, that's. They were the. For one of the first publishing companies. And you'd really anticipate, be really excited every. Every month to get their. Their cassette.

And so they bought my software. I was very excited, and I was very much in business.

And then finally, when I was able.

To afford a car, which was, you know.

Maybe six months later, I was.

Able to drive to another company.

In.

Gilroy, which was a much farther, which was a pretty far drive for me.

And I became a real programmer. And so when I was 16 and I've been working on software ever since, it is amazing.

And I want to transition to AI in a second, but you taught me the concept of a beginner's mind. And you talk about the beginner's mind all the time. And I thought, I'm going to experiment now and I'm going to channel Mark Benioff. So I went to Lovable, which is an app, and that. That app creates app software. And I put in the command, can you create an app that teaches people how to juggle? And guess how long it took?

I really have no idea, but I bet it wasn't very long.

Two minutes.

Wow.

Isn't that great?

Amazing.

So we live in the age of a marvelous age where everything is going to be transformed through AI, every sales process. So can you tell us a little bit about AgentForce? How did the idea germinate? How did you pivot just before Dreamforce to make this the key theme? And how did you go to Agent Force 2.0? And. And now I also want to know what's on the horizon.

Well, you're right. You have to cultivate a beginner's mind. And that is one of the key things, especially at this point in history. The Japanese use the word shoshin, shoshan, meaning a beginner's mind. It's a kind of a Zen construct. And this idea that we have to kind of look at everything new. And of course, in technology, that's the only way you're ever going to succeed anyway, because technology is always getting lower cost and easier to use. It's a continuum. So this is just at a moment in time where we are seeing incredible new technology, especially with artificial intelligence. And this is a moment where all of us are really going to see some incredible things.

Now, we already have a lot of.

That on our phones, right? You probably have one of these dozen AI apps where you can ask it all kinds of questions and things associated with what's going on in the world. And it reminds me a little bit like when I had an Encyclopedia Britannica, when I was like, wanted to know something, I would have to go on my bookshelf, grab my Encyclopedia Britannica. That's when I would be able to, like, learn what was going on. Now instead, I just use the app and ask the question, hey, what about this? What about that? I want to learn about this. Compare this to that. You know, give me. Explain this to me. It's. It's incredible. And I use these AI apps instead of using, you know, search, you know, kind of the traditional search box. So it's really created a new paradigm in only a few years for all of us. And the other thing that's really cool about this AI is that the AI that was the top of the game maybe six months ago is not the top of the game now. So it's kind of a horse race. And they're all kind of trying to compete with each other. And yet all that's in open source. So if you want to build your own AI, you can grab it off the open source and make it happen. Well, all of this is creating something amazing, which is going to be a revolution in digital labor. And this digital labor revolution means that what we had to, you know, have human labor to get things done, especially in companies like what you mentioned, salespeople or service people or marketing people, on and on and on. To have a company, to build a company, to create a company, to execute a company, yes, you're going to have human beings because we're the source of a lot of these ideas and inspiration and creativity, but also we're going to be augmented with agents. And we see it already at Salesforce. And I've got some great story, but this idea that these agents are very much digital labor, and soon robots, also an extension of digital labor. That this idea that, you know, humans with agents are creating customer success together.

Tell me a little bit about that. Initial, you know, in preparation for Dreamforce, you were on a different track. And then at what point did. Did you realize that agent force has to be front and center?

Well, the cool thing is that now, you know, when we've kind of introduced this concept of, all right, we're all going to move into this new world of digital labor. This is, you know, what software, Gerhard what's been exciting about software is that it's been a, you know, multi hundred.

Billion dollar tam, let's say.

That's why, you know, Salesforce has said.

We'Re going to do about $38 billion in revenue this year. So that's very cool because we're operating within a TAM that's hundreds of billions.

Of dollars for enterprise software.

But there's something big going on which is that when you think about this digital labor, tam TAM is total addressable market. This is much bigger than enterprise software. This is much bigger than just managing and sharing information for customers. This is the biggest, most exciting market really in the history of business. This is the idea of this digital labor market. So as an example, at Salesforce we have about 9,000 people dedicated to doing customer support. And that's not an unusual thing. We have about 75,000 people in the company. And so we built a site called help salesforce.com now that site is connected with our service cloud products very, you know, it's a critical part of our infrastructure, our data cloud. And then we built an agentic layer on top of that. AI agents. And that agentic layer, well, what's interesting about that is it is really productive. So what was maybe thousands of customers or tens of thousands of customers or hundreds of thousands of customers who had to go to support before, now they're able to actually get their questions answered by the agents. And when they can't get their questions answered by the agents, it's instantaneous. They're able to then go right to a human who has all of the information on the screen summarized for them to by the agent saying, hey, I got stuck here. I don't know what to do. I'm an AI agent. I can only go this far, but I did this. So that is what's very exciting is that humans with agents can work together. And at Salesforce, well, we'll probably be able to take now 4 or 5,000 of our employees who were working on customer support and move them to other critical functions like sales, you know, or things that require, you know, human interaction. And that is, wow, that is really different because even a year ago we could not do that and today that's working.

How much training do those agents get? How much data do you feed their brain?

Well, the real exciting thing is that we took our Customer360 product line, which is our sales product, you know, our service product, our marketing product, even products like Slack Tableau, which is our analytics capabilities, our commerce offering, our platform, all of our apps, and we rewrote them into a common core. And this common core is kind of a unified application offering. And we are continuing to enhance and grow that and extend it. And it's very, you know, it's very exciting for us to kind of have this, you know, fresh code base with the application applications. But then we extended it one more level up with a data cloud. And that data cloud then kind of.

Takes all the data up out of.

The apps and brings it together into.

A unified place to keep all the data.

But that data cloud does one other thing.

It federates, it does zero copy to.

Every other piece of data in your enterprise.

So if you're using databricks or you're.

Using BigQuery or using Snowflake or you're using Redshift or whatever you use IBM mainframes, whatever you're using, we're going to go out and connect to that and.

Then start to bring that into the data. So you have your applications where the humans are, right?

Then you have the data and then you have the agentic layer. And that agentic layer combined with the.

Data cloud, combined with the apps. It's really about apps, data and AI agents together that create this incredible new capability. And so we've seen some awesome stories now, I told you the Salesforce story, but there's another incredible company called Lennar. And what Lennar is this. Do you know who they are? They're the number one home builder in the United States. Yeah. So they're, they're based in Miami. They're an incredible company, multi generational company. And they have always had issues that they've wanted to transcend in their business that they've not been able to transcend. And simple things like number one is they want to provide 247 care to all of the homes that they've sold and all their homeowners. They want to be able to sell new products to all of those homeowners.

Like mortgage mortgages or insurance or these kinds of things.

They want to be able to do things with those homeowners and provide a level of expertise that extends beyond their employees. So what they did was they were a dreamforce because they're a huge Salesforce customer. And they went back to Miami and they did a hackathon. And this hackathon was really cool, where they came up with all these Use cases of how the AI is going to transform Lennar, and then they build it and then are deploying it. And the financial results that they're seeing, you know, they think they're going to save huge amounts of money by doing this, but they think they're going to make huge amounts of money also. And that is what I'm so excited about with so many of our customers.

So do you call this Agent Force 2, or do you call it Super Agents? What is the terminology here?

Well, this Agent Force platform has already had a couple of versions. I think we've gone from 1 to 2 to 3. You know, 3 is coming, 4 is coming, 5 is coming because it's going so fast. But when you see the customer success and what's happening, it's really incredible. I mean, another really great story is Singapore Airlines. I mean, everybody knows Singapore Airlines. It's a, you know, you're. You're amazing how you fly and all this. They do this incredible level of customer service. And, you know, they're, they're. They're known for their service. Everything from the time you buy your ticket to the time you're on the plane and how they treat their customers. Everybody knows it's one of the best airlines in the world. And so they were at Dreamforce as well, and they got the. They got the vision that they could take Agent Force and improve their customer service to even another level, which is kind of hard to imagine because they are so great and at what they do. And, yeah, they are deploying customer service. And what was cool is it was led by their co Go who, you know, he really got the vision, and he's an incredible person. Really had a great conversation with him about his vision for this, that he was going to be able to extend his, you know, airline with digital labor. Of course he's going to still have his pilots. He still has flight attendants, he still has people, people at the counters. He's still welcoming you at the airport. But that whole idea that you could provide a level of capability that is beyond even what all of his incredible employees are doing, and to that point, even in the airports themselves. Heathrow Airport is one of the first Agent Force customers that when you get to Heathrow, you have an AI app that is going to help you get to the right gate and make sure that you get to the right, you know, organizational, you know, point for you. But the thing that's interesting is, is that these aren't just AI agents that are just having kind of some transactional conversations, and then they move on. All of this is kept in these. In these apps and in our data cloud. And at any point while you're working with them, if you want to be connected to a human, you know that link is right there. Because it's about the apps, it's about the data cloud, it's about the agents all working together, because all of that really is just one piece of code. And this is what's getting me excited about what the possibilities are for the future, especially when I see so many stories emerging from our customers so, so quickly.

So you. You have published, like, three books now?

I think I've had a few books, exactly.

Could you technically put all those books into an agent brain and have people have a conversation with your agent and answer questions about the books?

Well, I think you could do something even bigger. You could take all the podcasts and you could take all the articles, and you could take all the books, and you could take everything and put it into a data cloud. But to that extent, let me tell you something cool that I just did at Salesforce. I just took all of it and put it into our data cloud, and I put an agent on the front of our website. So if you go to salesforce.com you'll see there's a little agent hanging out there. And that agent is connected to everything inside our company that we've kind of put jammed inside to our data cloud. And yes, it's all connected to our apps. And. And it's a whole different experience on. On the web. And I'm very excited about that.

You're telling me stories that. That really resonate two ways. One is they resonate because your agents help businesses grow. But the second part of the story is that your agents also can be deployed to create a higher touch experience, a better customer experience.

So where do you see the future? Where do you think this is going? Because when I look at technology, it has side effects, like medicine has side effects.

And.

You were probably the first one to talk about trust and accountability and guardrails. So can you share how sales leaders and CEOs need to think about the integration of agents with their workflow, but also having boundaries that people use it for good.

Well, I think that you're really touching on something that's really critical, which is that this is software number one. This is not magic. So this is software, and these are apps. This is a database, and this is agents AI. And it's all one piece of code at Salesforce. And in that platform, you're right, we have the capabilities to create the guardrails, the trust rails. The idea is that, you know, you're going to have the governance to make sure your agents are staying in the right lane, doing what you want them to do, not something else. And this is going to be very critical. And this isn't just some independent piece of software running outside of everything in your company. This is all deeply integrated into what you're doing. And that will be very important because you're right, this will be about your brand. You can believe that when we talk about companies like Salesforce or Lennar or Singapore Airlines, that these are big brands and those CEOs like me want to make sure that you're having a good experience with that agent and that you're getting your questions answered and that you're having a good service experience, you're having.

A good sales experience, you're having a.

Good marketing experience, whatever it is.

And we're at the beginning of this journey.

So today when you go to Heathrow Airport, for example, and you get to the ticket counter, there's not a robot yet behind the ticket counter that you're interacting with yet, right? It's still, there's human beings, but we can all imagine, because we've seen the movies and we know kind of where I think technology is, that within some kind of short order there's going to be some kind of robotic layer as well. It's not just going to be the self service machine that we have now, which is kind of the beginning.

You know, it's going to be more.

Of a humanoid type robot experience and it's going to still they'll be connected to your corporate database, it's still going to be connected to that guardrail system, it's still going to be connected to those apps and it's still going to be connected to humans because technology is not good or bad, it's what we're.

Doing with it that matters.

And this is going to take a while to get this exactly right.

But we're on the cusp of a.

New, a new vision for what's possible with technology.

And, and we can see that you can go to help.salesforce.com and see the beginning of that. And I think for many of our customers, like we have this great company, remarkable that makes these incredible digital pads.

And they have this running on their.

Service already or the Equinox gyms and they have it running on their service already. And so we're all kind of going to get into this in the beginning and I expect thousands and Thousands of companies will kind of really be out there this year and this is going to happen much faster than what we worked on initially together. You know, you probably remember 20 years ago we started working on the cloud together and you know, that was, there was a lot of hesitation. And you remember all of those conversations we had. Is this really going to happen? Why is this going to happen? And then we got into social, we got into mobile, we got into the beginning of AI predictive. That was 10 years ago. And now we're in the generative world and there's more AI coming, you know, in the future. So we.

This is what's exciting about being in the technology industry.

There's something else that I really like about your website is trailhead. Because people can go there and learn for free.

Absolutely. Well, I think this is really important. The reskilling, upskilling. Everyone is going to have to learn. We've been saying this now for a long time, you know, that trailhead is not something new. And we try to train and educate and create trailblazers and now we're going to try to create agent blazers and.

This idea of a huge, well educated, value based community. That's something that we're very much in favor of. And look, if you go back, Gerhard, the very first conversation we had was about philanthropy. We put 1% of our equity, our profit, our time into a foundation. That was our very first conversation. You were shocked. And now you look at, we've done a billion dollars of grants, we've done 10 million hours of volunteerism, we've done 50,000 non profits and NGOs run on.

Our service for free.

We've always been a values based company. Now of course, things are always changing, but our core values have not changed. We're still about trust, we're still about customer success, we're still about innovation, we're still about equality, we're still about sustainability. We, we're about all of these things and yet we're moving forward because we are a technology company now. We're now the second largest enterprise software company in the world after Microsoft, which is kind of hard to believe.

Right?

It's kind of amazing. Isn't this an amazing moment? You know, so it's fun to talk to you after a couple decades. You know, I don't know if we have recordings of our very first conversations, but they were very, you know, very. Not so different than this one.

Exactly.

I think that that's the key, Gerhard. Really, I really do think that is the key.

Well, you have influenced so Many leaders. And I'd like to pick your brain a little bit about leadership. What are the top three qualities of a successful leader? Not just.

Ordinary success, but consistently successful leader?

You've got to get clear about what you want. We already use one Japanese word here, which was shoshin. That's the beginner's mind. Well, that's beginning, process to, you know, you want to clear your mind, and then you want to say, what do I really want? You know, that Japanese word is ishii, intention. You know, what is your intention? You know, get crystal clear, have radical faith about what you want. And you've got to believe that you know, it's getting created whether you realize it or not.

So you better get clear.

What is it that I really want? What am I really trying to achieve here? Because that is going to happen. And, you know, what do you want for yourself, for your family, for your business, for the world?

What do you really want? That is, I think at the beginning of every hero's journey is what is.

The absolute intention and clarity that they need to have on what they want right now?

And the second part is, what we talked about earlier is then what is really important about it? What are the core values? What are those values that will guide that vision? Because you can achieve business success or any kind of success, life success, in a lot of different ways. But you have to decide what your core values are as well. And that's a whole different ballgame. And look, none of us are perfect. We all need a lot of forgiveness in our lives. We all know that.

But I can tell you this, that.

That idea that.

We can adjust our core values and kind of get to a greater level of success. This is really important.

What did you learn from Tony Robbins?

I learned from Tony Robbins that it's that key part of your state that's important. You've got to get in the right state, the right state of mind, the right state of body. You've got to really be ready in your heart of hearts to know, okay, I'm going to create the success because your state is connected to your success. Now also, the quality of your questions is connected to your quality of your success as well. You know, you can say that idea that asking a key question like, what do I really want? Is going to be really important to you. We can see that with the AI Prompt engineering is the quality of your question. You know, you don't put in the right question in the AI. You don't get the right answer out. And that's how life is as well.

That's what Werner Heisenberg said. Nature does not reveal its secrets. It only responds to our method of questioning.

That is very much deep down at a level of quantum physics. That's beyond my expertise, Gerhardt.

I know well beyond mine too.

I will not have a nod to my, my cousin Paul Benioff who was at Caltech who was I think probably one of the foremost thinkers on quantum computers and really came up with the initial concept of a quantum computer. That idea that you know that the state is very important and we're seeing so many breakthroughs right now in quantum. It's incredible. This is definitely going to be one of the next big areas that we're going to see really start to achieve. And your question will be answered.

Well, I want to commend you on your success. You have been at it for over 25 years and you still have the beginner's mind and you still have that, that mindset of that entrepreneur that is wide eyed, that is curious and want to see how to get through the next breakthrough.

And look Gerhard, we are going to be together for this digital labor revolution just like we were together for the cloud revolution and so many other revolutions. And I'll tell you that this digital labor revolution is going to be the biggest, most exciting revolution any of us have ever seen.

And I recommend anybody who is listening to us now head over to salesforce.com.

Check out trailhead but also check out the other website, what is, what was it?

Help.salesforce.com or Agent Force. Absolutely. Or just contact me directly on Twitter and I'll be happy to help you.

Thank you so much Mark.

Great to see you.

Bye Gerhardt.

Bye bye.

Bye now. Aloha.

Aloha.