Abhii Parakh

And I think with agentic AI, we have that opportunity where the service agents and the digital agents in whatever modality we choose, whether it's voice or text, but you can have one conversation and then the army of agents can coordinate, have the context on all the previous conversations with this customer, have that memory and also have the autonomy to be able to then deliver on the outcomes that they're after.

I mean, I think the technology is a little bit over promissory, if I'm being honest at this moment. Every company last year overnight became a Gen AI company. And every company this year has overnight become an agentic AI company. So while I appreciate that everybody's embracing technology really quickly, I just have not seen... the agentic side of things delivering on the promise so far.

I mean, I think the technology is a little bit over promissory, if I'm being honest at this moment. Every company last year overnight became a Gen AI company. And every company this year has overnight become an agentic AI company. So while I appreciate that everybody's embracing technology really quickly, I just have not seen... the agentic side of things delivering on the promise so far.

I mean, I think the technology is a little bit over promissory, if I'm being honest at this moment. Every company last year overnight became a Gen AI company. And every company this year has overnight become an agentic AI company. So while I appreciate that everybody's embracing technology really quickly, I just have not seen... the agentic side of things delivering on the promise so far.

A lot of that also has to do with how perfect the data has to be for that to happen, right? With Gen AI, I think it could happen much faster because it can be a point solution. You can point it at a PDF or a huge library of research or scripts that you have within your company, and then it can generate based on those patterns. But with

A lot of that also has to do with how perfect the data has to be for that to happen, right? With Gen AI, I think it could happen much faster because it can be a point solution. You can point it at a PDF or a huge library of research or scripts that you have within your company, and then it can generate based on those patterns. But with

A lot of that also has to do with how perfect the data has to be for that to happen, right? With Gen AI, I think it could happen much faster because it can be a point solution. You can point it at a PDF or a huge library of research or scripts that you have within your company, and then it can generate based on those patterns. But with

agentic AI, there are so many different systems and data points that have to connect with each other that I think That is the biggest challenge that I'm seeing for these agents to autonomously be able to go and get the task done. So I think we are looking at maybe another one to two years before this really becomes commonplace. I think there will be point solutions.

agentic AI, there are so many different systems and data points that have to connect with each other that I think That is the biggest challenge that I'm seeing for these agents to autonomously be able to go and get the task done. So I think we are looking at maybe another one to two years before this really becomes commonplace. I think there will be point solutions.

agentic AI, there are so many different systems and data points that have to connect with each other that I think That is the biggest challenge that I'm seeing for these agents to autonomously be able to go and get the task done. So I think we are looking at maybe another one to two years before this really becomes commonplace. I think there will be point solutions.

which you can point for specific things that live in a one-dimensional space. If all your data, all your stack is in one system of record, then I think it's going to be much more effective. But for truly autonomous systems where the customer can ask anything and they can go through different places, that orchestration, I think it's going to take some time for the kinks to be hashed out.

which you can point for specific things that live in a one-dimensional space. If all your data, all your stack is in one system of record, then I think it's going to be much more effective. But for truly autonomous systems where the customer can ask anything and they can go through different places, that orchestration, I think it's going to take some time for the kinks to be hashed out.

which you can point for specific things that live in a one-dimensional space. If all your data, all your stack is in one system of record, then I think it's going to be much more effective. But for truly autonomous systems where the customer can ask anything and they can go through different places, that orchestration, I think it's going to take some time for the kinks to be hashed out.

I would say that some of the Gen AI stuff, especially when it comes to being employee facing, but also customer facing is ready, right? We're doing that. It's ready. We're already in the middle of that phase.

I would say that some of the Gen AI stuff, especially when it comes to being employee facing, but also customer facing is ready, right? We're doing that. It's ready. We're already in the middle of that phase.

I would say that some of the Gen AI stuff, especially when it comes to being employee facing, but also customer facing is ready, right? We're doing that. It's ready. We're already in the middle of that phase.

I think with agentic AI, if you want to, this is the year that you should be trying and experimenting with low risk use cases, which mostly might be in the form of internal employee assistive use cases. But if you have low risk use cases, which are heavily content oriented, for example, you have a lot of content about your products and your products are complicated.

I think with agentic AI, if you want to, this is the year that you should be trying and experimenting with low risk use cases, which mostly might be in the form of internal employee assistive use cases. But if you have low risk use cases, which are heavily content oriented, for example, you have a lot of content about your products and your products are complicated.

I think with agentic AI, if you want to, this is the year that you should be trying and experimenting with low risk use cases, which mostly might be in the form of internal employee assistive use cases. But if you have low risk use cases, which are heavily content oriented, for example, you have a lot of content about your products and your products are complicated.

You might have an agentic AI system that may be able to go through that, answer questions, but also then connect you with the right people autonomously and maybe even ask you questions and give you some answers around financial calculations, all of that being in the realm of guidance and not advice, that could be an interesting one where you can do it with low risk.