97% Would Switch Vendors for Better AI Ops—Here's Why

Bob, Principal Analyst at Cube Research, reveals fascinating insights on enterprise AI adoption from his latest research. His four-stage adoption model shows organizations progressing from basic alerts (21%) to recommendations (28%), manual automation (26%), and fully automated responses (24%), highlighting the journey as enterprises build trust in AI systems.

• Research shows enterprises slowly gaining comfort with AI adoption across four distinct stages
• An astonishing 97% of enterprises would switch vendors for better AI operations technology
• The 20-60-20 "maturity model" applies to AI adoption with leaders embracing full automation
• Getting started with AI is crucial for organizations to build experience and confidence
• Consumer AI applications like ChatGPT help enterprises get comfortable with the technology
• Network complexity increases the need for AI-driven operational efficiency
• To identify genuine AI solutions, look for vendors with 5-10+ years of experience
• Effective AI systems incorporate closed-loop feedback and leverage customer support insights

-----

Key points covered: 

Adoption of AI in enterprises: The discussion highlights the different stages of AI adoption in enterprises, ranging from generating alerts to fully automating responses. It emphasizes the importance of organizations becoming comfortable with AI technology and gradually transitioning from manual automation to full automation. 

Impact of AI on vendor selection: The conversation reveals that a significant percentage of enterprises are willing to switch vendors if they can get better AI operations technology. This underscores the value enterprises place on AI solutions and their potential to drive operational efficiency. 

Evaluating AI solutions: The document discussion provides insights into how enterprises can differentiate between marketing AI and real AI. Key indicators include the vendor's experience, the presence of a closed-loop system, and collaboration with customer support teams to ensure the efficacy of AI solutions. 

-----

Where to find Bob Laliberte? 

LinkedIn - https://www.linkedin.com/in/boblaliberte90/ 

Website - https://thecuberesearch.com/ 
 

Where to find Bob Friday? 

LinkedIn - https://www.linkedin.com/in/bobfriday/  

-----

Keywords:

AI adoption, comfort with AI, AI vendor, closed-loop system, customer support, collaboration

-----

To stay updated on the latest episodes of The Q&AI Podcast and other exciting content, subscribe to our podcast on the following channels:

Apple Podcasts - https://podcasts.apple.com/us/podcast/the-q-ai-podcast/id1774055892

Spotify - https://open.spotify.com/show/0S1A318OkkstWZROYOn3dU?si=5d2347e0696640c2

YouTube - https://www.youtube.com/playlist?list=PLGvolzhkU_gTogP5IBMfwZ7glLp_Tqp-C

We hope you enjoyed this episode!

Chapters

• 0:02 Introduction
• 1:03 AI Adoption Research Findings
• 3:28 Getting Comfortable with AI Technology
• 4:34 AI's Impact on Vendor Selection
• 5:41 How to Identify Mature AI Solutions
• 6:54 Key Takeaways and Closing

Transcript

Bob Friday Host 00:05 

Hello, welcome to another episode of Bob Friday Talks. Today I'm joined by Bob, Principal Analyst at the Cube Research, and today we're going to be looking at the adoption of AI in the enterprise. You know, Bob, welcome. You know, maybe start with a little background on yourself for the audience.  

Bob Laliberte Guest 00:19 

Yeah, thanks very much, Appreciate it. So I'm an industry analyst. I've been covering networking for about 15 or so years 35 years in various industries, including optical networking and places like that, but really have a real passion for where the networking space is going and the innovations that are helping to drive it.  

Bob Friday Host 00:39 

Okay, so we were talking a little bit about you know chat, gpt, you know self-driving cars and everything. It seemed like we're past this skeptical phase of AI in most enterprises, so I thought maybe you'd give the audience a little bit about the research you're seeing on. Were these enterprises in the adoption of AI?  

Bob Laliberte Guest 00:56 

Yeah, that's a really good question. I mean it is interesting because I've always had the theme of time to comfort with AI. Actually, any new technology right, it just doesn't matter. But AI is so transformational maybe it's gonna take a little bit longer.  

01:09 

So for the last couple of years I've been asking a question in my research to find out where organizations are in basically a spectrum of uses of that, and so the way it breaks down is typically I've asked three responses. I've had three responses. I've had it say I want AI to generate an alert to let me know something's wrong. I want AI to generate alert and give me a recommendation of how to fix it, and then I want AI to generate the alert right and then take that recommendation and fully automate the response. Now this last year when I did it the very last one I always thought that there was a stage in between where it was give me the recommendation and then I'm going to manually automate the response before I go to the full automation. So this last one I did, I asked it in that four-part response and what I got was 21% came back and said just give me the alerts, that's all, I'll go fix everything.  

02:06 

28% came back and said I want to have an alert and recommendation and then I'll go fix it, but obviously it's going to be expedited. 26% came back and said I want the alert recommendation and I'm going to manually automate the response, and 24% said I want to have it fully automated. So I think we're definitely. The things that I've seen by splitting it up is that that middle section obviously has been always the biggest part because of the conservative nature. They need to be able to understand and really check the efficacy of the response that they're getting and validate it themselves. But then, as they become comfortable, you know, I always ask right, what's the transition? How long does it take to get from one stage to the other, and how many times do you have to do the manual automation before you'll go to fully automated, and can you do that for certain responses? So instead of going fully automated for everything, it's fully automated for the ones that you've fully vetted.  

Bob Friday Host 03:04 

Yeah, that sounds very similar to my self-driving Waymo Uber. When I asked my friends, I would say about 25% of them will get in the car. The other 75% are not quite ready to get in the car. Right, yeah, so it sounds like we're thinking like 25% of enterprises are ready to hop on board their self-driving network venture.  

Bob Laliberte Guest 03:29 

Yeah, and I think, like I said, it's always going to be organizations who consider themselves to be, you know, leaders and right, that's. That's a classic maturity model of the 20, 60, 20 is kind of the leaders, the followers and, I'll say, the more conservative people right that are out there and so. But what it takes is experience and so, as we've talked about before, it's really important. The most important thing for organizations as they think about this journey to AI is to get started so that they can actually work with the product and understand it, and, in some ways, that's where the commercialization of a lot of these AI technologies chat, gpt and things like that have been incredibly helpful and things like that have been incredibly helpful. So organizations can start playing with it, getting comfortable with the concept of AI, and then thinking, okay, how do we bring this in to my network environment? Because it's certainly, as we all know, it's getting a lot more complex, right, so they need help to drive that operational efficiency.  

Bob Friday Host 04:18 

And we talked a little bit before the show. It was like where do you think AI fits in the priority of vendor selection? In terms of nowadays? This is a paradigm shift of hey, in addition to keeping all these things green, we now have kind of this client to cloud AI component to try to keep the whole experience flowing.  

Bob Laliberte Guest 04:34 

Yeah, it's really interesting Some of the research that we did. We actually we asked this a lot. We'll always ask, hey, if this new technology comes out, are you likely to switch because of it? Technology comes out, you know, are you likely to switch because of it? And typically we'll get somewhere around, you know, 50% for a really kind of snazzy new technology. When we asked this question and said, hey, if, if you had a vendor with better AI ops technology, how likely would you be to switch and go to a whole new vendor which, as we all know right, that's a big jump going from one vendor to another 97% of the respondents said that they would be willing to switch if they could get better AI ops. So I think that's a really across-the-board recognition that they understand the value. It might take them some time to get there, but they know if they've got a better AI solution that they can leverage, it's going to drive real benefits for that organization.  

Bob Friday Host 05:24 

Well, Bob, that brings joy to my heart. Maybe to kind of wrap it up. And finally, what is your recommendation to enterprise customers who are trying to sort out the marketing AI from the real AI? Do you have customers or enterprises asking you to help figure this out?  

Bob Laliberte Guest 05:41 

Yeah, that's a really great question and, because of all the AI washing that's going on, it can get really confusing because everyone's saying they've got it. So some of the key things I always tell them to look for are how long have they been doing this? Because most of the ones that are really gaining traction today are ones that have been around five, 10 or more years. Right, where you've got the data, you've got the models built up, you've got the efficacy built up so that you're delivering real positive results.  

06:10 

One of the other questions I always make sure that people ask is make sure that there's a closed loop system as part of it. Right, is the vendor actually concerned about getting your feedback? Because you've got people with 30, 40 years of experience out there who know these systems inside and out, and so the ability to leverage their knowledge is only going to make the AI systems better. So, having that closed loop and then another key one that I usually point out to is make sure you ask your vendor if they're working with their customer support teams, because that's where all the problems if you're trying to solve problems, that's ground zero for where they're at. So those are some of the things that I usually indicate to organizations, say look for these things to determine whether it's a mature AI, ops solution and program.  

Bob Friday Host 06:54 

Okay, well, bob, I want to thank you for joining us. Totally interesting. Thank you audience for joining us and look forward to seeing you on the next Bob Bright Talks and check us out on the Q&AI channel.