Ep. 2: OpenAI, Google & Gartner: What’s Happening in Voice AI

Welcome to Bespoken AI podcast. Press one for the latest news in voice AI. Press two to start the show.

Hello, welcome to the Bespoken AI show. My name is Diego, head of business development at Bespoken. Uh joined here with John, our co-founder and CEO.

Nice to be here. We're very excited to do this first podcast live in Barancco. So that's pretty cool.

Uh that's Barono in Lima, Peru. and we've got uh Peruvian coffee. We've got good stories and we're going to cover a few headlines that we're uh excited to talk about uh covering voice AI enterprise and a lot of Google today. So, we [laughter] we we'll be talking a lot about what's happening. So, let's jump in. Uh we've got a first story. Um this is a headline that we saw on TechCrunch. uh talks about a open AAI launching a uh the way they say it is a screen-free device screenless uh portable and AIdriven. So we're going to suspect that that's a very heavy voice AI product. Uh it was developed by Johnny IV or as the UK refers to him as Sir Jonathan Paul IV uh once he was kned. And this is going to be a donut-shaped uh speaker. It's going to be around $200 to $300 uh dollars.

And it reminds us a lot about the Amazon uh Echo Dot. So, we're going to pull that one up in the screen because that's the closest thing we have uh as a reference. [snorts] And um yeah, uh curious to get your thoughts on that, John.

I think first of all, this is really an indication that uh Chachi PT and Open AI are very serious about voice. Uh so that for me is a great sign. Uh it's a reason to be excited. Uh I think it shows that they're going after to some extent uh Apple and Amazon. Uh because this looks like a product that is somewhat competitive with things that they offer or have offered. So on the consumer side, I think it shows a really aggressive strategy. And then in addition, and this is us reading into it a little bit, I would say, but uh you know, I think it shows that they're continuing to go after Sierra and ElevenLabs and that they want to be very involved in the enterprise and that goes to the technology that's really behind this. Um so, you know, the fact that Open AI is so bullish about the voice AI space, I think it makes us

feel excited about what we're doing, uh and what everyone's doing in our industry. Uh, and I think also especially so perhaps because they have done this calling of products. They've gotten rid of a lot of stuff. They got rid of Sora, uh, their video product that they're doing with Disney. Um, you know, they've gotten rid of a number of other initiatives. Uh, the fact that they're still really doubling and tripling down on voice, I think is fantastic.

And a few additional notes on the context. Uh Johnny IV is the former chief design officer of Apple and he is credited with uh developing OpenAI's first hardware product. So this is going to be the first hardware product uh related to Chat GPT that we're going to see. Um and I don't know if there's any other context. Well, one important note is that I think OpenAI got a lawsuit with Apple. I think Apple was accusing them of of hire like hiring all of these Apple engineers and made the case that you know

and Joanie was supposedly telling people like take your stuff when you leave the building.

I didn't read that but but

or maybe so people that had left Apple. Yeah.

Yeah. So so I but I think you know Sam Alman is going to make the case that you know this is uh a very new product

something like we haven't seen before. So that's going to be really interesting. But uh you've you've you've got a firsthand look at the Amazon uh AI voice speaker wave uh back in in the voice wars, right? So that was around 2017.

Yeah. So I mean I I think that that part uh I mean that's a really interesting angle with this which is you know we've got the Echo device up here. I you know I'm not sure this looks like the coolest thing right now. I mean we we were talking about this beforehand. The fact that they are uh calling it a portable uh screenfree device as opposed to a smart speaker I think is interesting. I think that also indicates maybe this wasn't a hugely successful product category. Um, obviously I think the Amazon Echo is an amazing device. Uh, or at least I, you know, uh, it was part of the reason why we started Bespoken was we wanted to help people build for this platform. But, um, I don't think it's been quite the success that everybody hoped for. Uh, so it is interesting to me. I mean, I'm I'm really happy OpenAI is doing this. I think that's fantastic.

Uh, I think, you know, their commitment to voice AI is exciting. You know, I I think myself and other people, you know, you scratch your head a little bit at this choice if this does turn out to be what they're going to put out there.

I'm trying to think. So, there's like a a bit of a resurgence, you could say. Uh, you know, because that was a wave in its own. It was like the voice wars. It there was also the Google version. Yep. Um, so do you think OpenAI because that that period of like I think it was uh by the time it was 2023 uh Amazon Alexa and and these speakers were losing like $25 million. Right. Right. Right.

So So what were they too early? Is this the next way? Like are they trying to pick up where they left off and and and make make the case that Open AI can do it better?

I mean I think that's a great question. They definitely can do it better. I think one of the the issues that Alexa had was the fact that it wasn't very intuitive to talk to. I mean, so many people would say, "Oh, this is good as a timer,

uh, as good as an alarm clock or white noise machine, but it's not good for much more than that. Even though there was a very deep ecosystem around it and a lot of people worked very hard on it,

it was hard to use it in an intuitive way." Uh, and you and I have both played around with the new chat GPT live

uh or chatgpt app where you know that's like a demo or where you can start to use chatbt live and talk to it and it's amazing. I mean it's it's an amazing experience as a user. It's intuitive. uh it has access to all the content um and the sort of depth of knowledge that uh chatbt offers and it really feels like you can just have such a natural conversation with it which I you know I think is amazing. So that part you know uh the smart speakers have a little bit of a bad name. I think they're a little tainted now but

you know would they have that bad name if you could have had that really natural experience talking to it? And that natural experience uh took a big leap recently with uh the speechtoech model. So uh let's talk about that for a bit. What what is speech to speech? Maybe for those that are hearing that for the first time. Um I you could maybe give like a small summary of what that what that's like.

Well, I so the speech to speech I think is amazing. So that is, you know, traditionally what we're accustomed to doing is you're going to do uh if you're working with voice AI, you're going to have a speech recognition uh step that's going to take audio, turn it into text, then you're going to pass it through a filter. It used to be an NLU, now it'll be a large uh language model. Then potentially another step in which you're going to uh you know once you get uh that response back uh you know come up with an actual textual response. So sort of a third processing and then finally send it back into audio and back to the actual uh

you know the person that was asking the question. Um

and that brought about like what we call latency which is like you're just waiting for a response for it's just unnatural right? So

that model has a lot of benefits in terms of control. There's a lot of accuracy. uh it does have some issues with accuracy as well, but the fact is people have been using those models for years and years and years. And so with the speech-to-pech model, you don't have that same level of control. You can't gatekeep or direct the conversation as it's going along nearly as easily.

Do you see any limitations on from a QA quality assurance perspective from the speech to speech? I think that it's going to take people a while to well I've thought people will take a while to adopt this. I will say now that we see the progress that OpenAI has made as well as um I would say just the general rate at which things are moving, I do think that we'll probably see people putting these into production in 2027.

Uh one because I think they'll overcome uh the ability to set guard rails on these, you know, very soon uh if they haven't already. even if it's not completely productionalized. Uh and then in addition, I think there's just going to be people who say, you know, what the heck, you know, this is this is safe enough because uh you know, I I'm not the only one who feels this way. The rate at which these models have progressed is is astonishing and I think it's obstruct most people's expectations. And I think noteworthy there is Open AI is investing heavily into voice AI and some other big names we're hearing is ElevenLabs, Sierra. Uh we're seeing are we seeing Open AI take a stab at them or is it is it uh you know

Yes, I think they are. I mean, and I do interpret this as their desire to take those guys on. Uh that said, I mean, you're talking to as many customers as I am. I mean what do you like do you hear anybody saying that they're using open AI in their systems? I haven't heard it.

I am hearing a lot of custom development of voice agents and I've maybe only come across one that explicitly said open AI but um yeah no I think you know these 11 lab Sierra platforms offering offtheshelf voice agents that's I'm seeing that conversation a lot more than let's say our traditional CCaaS uh Amazon connects. There's just a lot of talk about it. Um,

I agree. I agree.

So, it will be curious to see if OpenAI opens up that sort of contact center. Voice AI, here's your voice agent off the shelf. Uh, I don't think we've seen any major headline on that yet. But with the chat GPT live, I think that's a big step towards towards that.

We're planning on sharing it directly during this podcast, but we've set up a predictions website. Uh, we want to start to try to

try to track some of the different forecasts we make. And I mean one because I love to make forecasts and predictions and I do think and this is optimistic thinking but I think there's reasons to believe it uh just based on the facts as well but I think open AI is going to make more of a play for enterprise. I you know uh it's not the same as the coding market but I think they do feel like they lost a little bit of a step to anthropic and claude because they ignored enterprise coding

and so enterprise voice AI enterprise CXAI you know those those are the two that everybody mentions. I mean it's coding and CX right customer support. Um,

so I mean for OpenAI to say no, we're we're not going to just hand that over to Sierra and ElevenLabs,

uh, that would make a lot of sense to me and I I think we will see that in 2027, but um, who knows?

Yeah, in the next episode we'll have that app ready to

just go visit. We can revisit this.

We can post we'll do the social media post with this and maybe include some of those predictions.

That'd be cool. Um, so in this in this podcast, we we really want to share these stories, but also maybe share our gut reactions. And uh, when I looked at the headline from TechCrunch, kind of cringed a little because, [laughter] you know, they the way they they the captioned it was like screenfree uh, device that you can move around. And I'm like, that's such a weird way to sell it. and like and and so so it almost seems like they're not even trying to sell it. So what what we are also hearing are rumors that OpenAI could even build a phone and so that's that's a big story too.

Um so I'm trying to

keep track of that. Uh you know are they really going to launch this speaker? Is it a distraction? And the real story is an an OpenAI phone. Uh, I'm glad I didn't have too much coffee before [laughter] that because I'm violently enthusiastic for someone else to build a phone. Uh, I, you know, I like my iPhone. I really like Siri, which is, I think, maybe a little bit contrarian, but I do think there's so much room for innovation on the phone. I've been frustrated with Google. I don't think they do enough with Pixel. Uh, we're in Peru. You can't buy Pixel phones here. I find that, you know, like Google does not dane to actually sell the thing worldwide. It's just unserious. Um, and I don't think Samsung does a good enough job with software. So, to me,

it's just like Apple's been seated this market that is so enormous and it's just ready for someone to take it on.

Uh, I always wish Amazon would take it on. I thought the Echo, well, actually a phone attempt from them predated the Echo, which was notably fail a failure, but I thought, "Oh, they'll get back to that." They never did. I still don't really understand why uh though I've heard some different reasons but to me it would make so much sense for Amazon to do it and if not Amazon I hope open AI is ambitious enough because I think the opportunity is immense uh to be AI first with a phone obviously it's tremendous opportunity and I don't as great as Google is as incredible as a lot of the stuff they do is I don't think that they have that sort of desire or ambition to innovate for consumers

um in a way that Apple does or or just a new player might. So,

I hadn't seen that story. I mean, you pointed that out, but I I hope they do. I think it would just be just a wonderful thing.

Um, you know, probably not so good for privacy, but uh overall a wonderful thing.

So, so you mentioned Google. Uh Google is they featured an article about 59 uh acquiring a $100 million deal with a financial services company for CCaaS. Uh I think the the the key word there is CCaaS. U there's a lot of uh it could include AI and I'd love to get your reaction to that. Um it said you know it mentions that it could eventually involve voice AI agents. Uh so that's that's a lot of coulds we could talk about. Um and I think the real story though is just you know Google positioning itself in a way that it's bringing contact center business to to these CCaaS platforms. um you know it's highlighting its ability to streamline procurement billing um and that there's business and I think you've you've seen the Google marketplace uh maybe in its early days uh what what was that like?

Uh well the Google marketplace when we were using it uh was not I think the most serious business endeavor. I mean, I'm trying to be somewhat diplomatic in putting it that way, but it it was very difficult to get approved for. Uh I would characterize it as uh you know uh Byzantine and sort of uh just just challenging in every way just to get approved and then you know I mean we as a software vendor we do participate in a lot of marketplaces and have a lot of partnerships and you know you expect

yeah there's some bureaucracy and getting set up as a partner uh and approval processes as there should be um but then also there's going to be some commitment from the partner to promote you to kind of way to work you into their channels. Uh Amazon I think is the example par excellence for this. I mean Amazon um I mean it's tough. Um it's not like they just hand out money to you or hand out you know deals with customers but they have you know a very transparent well-defined sort of program by which you can progress and you can really build a whole business on AWS marketplace.

Uh and they transact billions of dollars there. It's it's amazing. uh Google does not do any of that in our experience as a company. Uh there was no promotion after we launched and yeah it was a little bit of a let down and we we came in there we were part of a Google startup program. So we were we had the impertor of Google saying this is a good startup and somebody was talking to we had Google high level people at Google saying please help them get live and we couldn't. Mhm.

Um, you know, that's just one company's experience with it. But I I think that that goes to one of the things I think is so interesting about the story, which is that, um, you know, 59, you know, they obviously see a different, you know, see the Google marketplace differently. They see it as a great procurement vehicle. Uh, they see it as a delivery vehicle for technology. And I think, you know, they they covered this in the story. um they see it as uh this is a way to you know ensure your data is safe. You know everything is going to be safely enscconced within your Google cloud environment. Uh and so just overall a way to really make the whole uh motion of bringing customers live so much easier. Um, so I, you know, I mean, one, kudos to 59, I think, in Google for doing that because I, um, it's a different delivery model than I was expecting to hear about this year.

Um, and then the obvious question that it also raises in my mind, uh, is, you know, does this signal a potential acquisition where Google maybe would buy 59 down the road?

So, I, you know, our experiences with the marketplace aside, I think it's it's pretty cool. Um, we'll see if we hear more about the cloud marketplace going forward.

So, you touched on an interesting thought there of uh why why would Google acquire a CCaaS and if it were why 59, but more importantly, what does Google have already u towards a contact center AI? Um the could we could you give give a bit of a background on what where

I mean I think this is this is where I do find Google's behavior often to be somewhat perplexing because to me it's so natural that they would want to have a see uh and when I say that what I really mean is they want to have a channel by which uh companies can talk directly to their customers. Why do they want that channel? because that channel I mean that's the pipe you can push all the AI down.

So it's it's such a natural delivery vehicle. Without that delivery vehicle, Google is left to go around to the 59s

uh the Genesis uh the nice uh in contacts and even the Amazons of the world and say hey would you adopt our AI technology? We think it's the best in the world. A lot of people agree with us,

but you know, uh it's much easier rather than trying to convince those companies to just sell your technology directly uh to the end customer.

Um whether it be the the companies implementing it or them, you know, then provisioning it to their actual users. Uh so I do find it uh surprising in a way that they haven't done this yet. I think that they toyed with the idea

uh in sort of the previous round of voice wars. So, I think that they, you know, it, you know, I mean, I I'm not privy to what goes on in the Google boardroom, but there just seem to be some signs that there was a couple companies they were looking at.

They didn't do it. But now, this announcement um you know, to me, it would potentially signal that they can

um you know, they're looking they're looking to get this channel and moreover that they want to keep up with Amazon because Amazon does have Connect. I think they're so well positioned. I mean they don't have all the great headlines about how great their AI is but they have connect which is a great channel.

And what makes Amazon well positioned to be like a winner in that space? Is it the fact that the data lives there or

Well, I mean that's a great point too. So I mean the data one I would come back to that but overall just I mean with connect um I mean Amazon has been investing in that for years. I mean, I think you and I found a chart saying it's is it number two now?

Oh, yeah. And that chart with uh Genesis was in the lead. Amazon Connect was was followed.

Yeah. I mean, there's so many great contact center companies and technologies have been around for a while. For [clears throat] Amazon to take a product that's been launched only relatively recently and already be in the number two position is amazing. Yeah, we'll have to find that graph and we'll we'll share it in the next episode too because that that was a a chart highlighting I think annual revenue by CCAS platform. So,

um Genesis was in the lead and uh some other players I think were nice and

yeah, nice and 59 I'm sure were on there

and um 59. So, they almost got acquired. they they almost got acquired by Zoom and and we we spoke with that team and I think what

well it's going to be interesting to see if you know if Google is a potential uh buyer that that would be um an interesting point in terms of you know being a channel for Google to apply the AI to end customers. So I think that's that that's a interesting point. Um what else? So 59 Google I think that's a good segue into the next story which is uh Gartner's magic quadrant and um

surprise surprise Google was in the lead

and u if you could show us the uh the graph here Google's in the far right of visionary

really

doing well here

the most visionary and the most leader [laughter] so that's interesting um we have some other notable names on Here we have PolyAI, we have Soundhound, Nice, Salesforce. Uh, [snorts] but there were some uh very big names we didn't see on this list. Um, primarily Amazon Connect, um, Genesis,

and even the newcomer, well, not newcomers, but the

I think I think a lot of the people's favorite, uh, ElevenLabs, Sierra, uh, they're not on this quadrant for conversational AI platforms. So we are going to talk about you know what was the criteria for this? Um what is it signaling and yeah I think uh

well the criteria I mean I I think I made some notes on this um but you know they said that they needed to have 200 plus customers uh geographic distribution and multi- channelannel support. Um,

and we're talking about, you know, Google potentially buying a 59 to have a conversational AI platform.

What does Google's tech stack look like?

I mean, I think that's a great point and these distinctions are probably, you know, of the many things with the magic quadrant that are suffering today with the rapid pace of change, you know, just these clean categories I think are one,

right? So I mean the whole idea that um you know the difference between a conversational AI platform and a CCAS

uh I think is going to seem kind of slight. I mean a CCAS has many components but I think the ones everybody agrees upon it has you know the bare ability the sort of networking to connect people via telephone, SMS you know email whatever your channels are

into agents. Uh it has workforce management which is huge. I mean for these people don't talk you know, in the AI space about workforce management, but that is the engine that really drives a lot of these products. It's just, you know, you've got a thousand agents, how do you deal with, you know, one just got fired, one just started, you know, one's on break, what are their hours, what are their languages, all sorts of crazy considerations that they have. Um and then you know the agent QA different from the type of QA that we do but making sure that you know uh the conversations people having are appropriate you know that they're on tone you know just generally making sure that your whole workforce uh is functioning optimally. So those those features are part of the CCaaS. The CI is okay. The part that we traditionally would call IVR, you know, or just a chatbot. That's what the CI is. This is the part where the robotic piece that

answers the phone call that annoys the heck out of everybody at the start of the call. That's what the CI platform is. And that's not only deployed through, you know, contact center channels, but most of these products, if you look at them,

you know, I mean, the lion share, not 90% plus of their revenue is going to be coming from a tiein to a CCAS or, you know, some sort of CX type of role.

I do see Soundhound uh in the leadership board. uh you know they come out they come out a lot uh and we we've we've even seen them on uh uh in the context of uh drive-throughs.

So that's a pretty fun topic. We we'll probably spend a whole episode talking about drive-throughs. Um and they're being used by, you know, uh major drive-throughs that, you know, they have a pretty good use case for voice AI. Um, so I I'm curious to know uh there's some other ones we haven't seen like Replicant didn't earn a spot on this graph.

Yeah, that was interesting.

So,

especially because I think they were on it last year. So, that was a little surprising to me.

But I I would like to explore Google a little more. Are are there conversational AI platforms that you have been, you know, involved in? Uh,

well, I just I mean, Google's inclusion way over here. I mean, this is You know, I'd like to have it illustrated. I would draw a circle around this and move them. Well, I mean, somewhere uh back here, probably down a bit. So, I guess, you know, they're trending this way. I don't know if I'm on the weather man here. you know, there's a low pressure zone. But I mean,

I just and it's not I mean I and if people listen to me talk about Google, I probably sound um you know uh not schizophrenic, I don't want to use [laughter] that word,

but I mean I have I have an intense admiration for Google. Their technology is always incredible but their execution especially their just their will to deliver to end users is always in question. I mean there is this feeling when you deal with Google that it's like are they actually doing things to serve the people using the software or are they just using it to you know doing stuff to impress each other and I a lot of times it seems like the latter and to this point I mean it's like yeah I mean the CI does not have to have a CCAS it doesn't have to have telephone connectivity

but you know completeness of vision and they don't have any way to actually use the thing on day one. I don't know. I mean, it doesn't feel that complete.

Um,

but it's also the fact that uh they've gone through so many iterations. I mean, they bought API.ai. Then it went through a major refactor to Dialogueflow uh with just plain Dialogflow. Then Dialogflow became Dialogflow ES and they put out Dialogueflow CX, which was, as far as I could tell, a completely different product. Just shared the same name and similar purposes, but really functioned almost, you know, underneath the covers was just completely disjoint from ES and I think confused the heck out of the market uh with that for a while. And now they've deprecated again where they have Gemini AI studio and again I mean on ability to execute I mean I I mean they can execute but have they when essentially Gemini AI studio is a product that's launched this year

uh and I believe it's still listed as in preview mode right

not even beta.

Yeah. So and and preview from Google is better than you know it may be higher quality than most software engineers ever build in their life but it's still you know it is it does correctly signal that it's incomplete you know we are users of a lot of Google's technology that's part of the reason why we admire them so much is that we know their stuff is good but if you want to use their new stuff it is scattershot throughout these different product names that I just went through and some other product lines as well. So, if you're looking for certain features, if you want streaming audio that uh or streaming text to speech uh that works really great, you may have to go to one place, but if you want it with a new model that has ElevenLabs type voices, you go to another place. And

you know, it's a mishmash. Um so, you know, I I won't go on too long in this, but it's just to say, I mean, is that really execution that far outpaces everyone else? I you know to me I would say no. Um and especially then when you compare it to you know like and then you know I'll just jump into the other part of this where the the other biggest mistake I see here which is Amazon's exclusion. I mean Amazon also has a little bit of a sorted history with uh deprecating products here. It does seem like with connect you know they've had a few bites at the apple in terms of what they're trying to do. I won't bore everybody with that story,

but you know, they've also had some churn, but it's like

I mean, they definitely have a stable kernel of product for conversational AI for the contact center in general.

And I just don't know what, you know, merits Google being up here up here and then Amazon just not even appearing. Um, so I mean those those parts to me are rather perplexing. um you know but and there's other points I can pick on too but before I go on too much

no I think I think uh I think we were surprised because before this magic quadrant we had we had been observing Google um and I think over the last you know month with all these headlines even even the uh 59 story the 59 story shouldn't have any impact on this because it's more of channel conversation than an actual capabilities. Uh

yes. Yeah, I agree with that. But I I will say I mean it's interesting you bring that up though because I do think that I mean that goes to okay I mean there's several things we can say about the quadrant and these analyses at a meta level and it's like if you're a contact center provider now uh you have to be providing a conversational AI platform and I think you know my point about Google acquiring 590 is like to me they sort of have to make an acquisition like that. I don't think that means that they will necessarily. I mean Google will do as they as they please, but to me it's like if you don't have that channel to deliver on, it's really not a complete offering.

Um so really to me it's that you need both. And you know 5'9 should start to enter the conversation and so then what does that do with these categorizations? Does it really sort of destroy a lot of these traditional barriers and categories?

I I think that's definitely something you're going to see with AI. I mean, we're already seeing it in our own business. It just it's so much easier to build software today. It's collapsing uh these walls that separate different, you know, traditional sort of lines of software.

And and to that point, I think that another connection with the 59 story is Google's positioning of like we've got the infrastructure, we've got, you know, Yeah. So, so we didn't touch on that point previously, but I think that was a big piece of uh the highlight of that story was like, you know, we'll be able to scale with the the AI implementation that you will need when you need it, you know, so that that was behind the

Yeah. and the fact that it's all provisioned within the sort of Google infrastructure. I thought that um well one I think intuitively it makes sense that you want to keep your data close to your AI.

Uh I think you know there's a lot of reasons why that makes sense just latency reasons. uh but then in addition I have seen enterprise teams really sell against that requirement where it's like look I mean even if our product and our algorithms are not as good do you really want to be shipping all your data to another vendor I mean we deal with a lot of customers um that are very sensitive about their data uh anything that's production data has any kind of PII in it I mean you really don't want to send it somewhere else so I mean that 59 you know with that announcement yeah I mean that is a nice aspect ect that they um [clears throat] you know essentially they can say look you you've got all this great functionality from 59 and then because you can leverage Google AI and other Google infrastructure you can keep it all housed in this one place and really I mean I you know I don't know what it's like to be the CIO of a Fortune 100 company now but I would really not you know like in terms of things that would cause me to lose sleep it's like just worrying about that headline in the Wall Street Journal that says Oh, JP Morgan, all your data just got exposed. It's like, oh boy,

you know, and so, you know, like what better peace of mind can they sell?

And and we are familiar. So, we've got some friends at IBM that IBM has been a really big partner for Bespoken. Uh we go way back. Uh but we've seen their conversational AI platform uh through some changes as well. Um, we don't have to spend too much time on that, but but you know, we've seen Watson. Uh, we've tested Watson a lot. Uh, I think it's probably worth noting their sort of position on this. Um, especially given now they're they're leaning towards more of an orchestration approach.

Yeah, I mean, I think IBM uh in previous iterations has shown higher. Um, and I think that and I I think it's welld deserved that they were higher in previous years. Um, because they do have a completeness of offering. Um, and I think this year, I mean, this position, you know, I probably would put them higher, but I think, you know, the ability to execute, I mean, I think that should really be labeled, you know, how are you executing?

I mean, IBM has this big challenge that they have Watson. it does have this mature great offering for the contact center, but they now need to retool it. And so they're incorporating ElevenLabs and they're trying to figure out how LLMs fit in. Um, and they're doing a lot of rework on their stuff. They're introducing control planes that you and I have talked about. You know, they're adding a lot of functionality to keep up with the AI stuff. And I think they're doing a reasonable job, but it is it's tough. I mean it, you know, anyone who's not, you know, and I guess this is if you want to make a case for why Google and [laughter] Amazon should be up here is the amount of resources you need to execute nowadays, it's just so enormous.

I mean, it really is.

Um, nice Cogni, they've got a good position here,

but

and I and I think everyone loves what Nice and Cogni are doing. they only get sort of plottits uh from people that I talk to but you know I mean they're not you know they don't have the resources of these guys

and this is not a predictive map this is you know sort of a snapshot of how things are today or as of June 2026. Um so what what what would be your take of how this map might look like next year? I would actually step back from that question to do more of a ma comment on Gartner. I think Gartner needs to retool. I think that this quadrant was a real misstep for them. Uh I'm still being somewhat diplomatic. I've seen commentary on LinkedIn that was just scathing. Um and I like look if you look at their criteria you can understand why they wouldn't include an ElevenLabs why they wouldn't include sir. I mean their thing is like look we're looking to jud judge people based on past performance

you know and you have to have this track record with customers it has to come through in our customer interviews I mean that's a reasonable way to root things obviously I mean it it makes perfect sense but it just you know it's one of these things where the world has changed with AI and it's like yeah that's I get it that's how we did things for a long time I have a lot of these conversations nowadays we we did things like that for a long time doesn't matter anymore

you know looking at vendors and their performance and what the product was like two years ago and that there was a successful implementation. I mean, it's not irrelevant, but it's not, you know, to say that Sierra and ElevenLabs don't have a place here is ridiculous. To say that all these sort of fairly narrow niches of software are not going to just collapse. I I I think that's also an assumption you have to question. um you know uh and then also you know they kind of keep well I won't point you know they keep some vendors around

that I mean let's face it no one you know I I don't think people are taking seriously at this point and so you know they're giving people based on looking backwards uh I think a longer play than perhaps they deserve and I won't name names there but I I do think uh

you know people like myself look at this diagram and say look some of these guys like they don't belong here anymore Mhm.

So, I think Gartner is going to have to find out how they stay relevant.

Um, they really are. And um and I don't think it means the magic quadrant necessarily goes away, but I think the criteria

uh perhaps changes and some of their methodology I think maybe needs to change too because the other thing and again I'm not endorsing these statements but you know there is always this sort of veneer that these things are paytoplay and I think when you you know see a bunch of vendors missing and a lot of sort of older more stayed vendors showing up

it opens you up to or you know it sort of invites that type of accusation probably unfair

um but nonetheless you know doesn't help Gartner's reputation and doesn't help the whole analyst industry quite frankly

and and we we haven't even touched on Salesforce but it's been coming up on conversations too with you know Agent Force and uh you know we we've been looking at their stock fairly recently but uh it's it's uh

yeah I mean Salesforce I mean maybe we'll have to edit this part out but Salesforce just makes me laugh. I'm sorry. I mean, if Agent Force is here on Completeness of Vision, why did they just buy Finn?

I mean, it doesn't make any sense. I mean, in theory,

I mean, I'm not sure. It looks like they would almost be the second highest scoring if you did some sort of multiplier

uh after Google, but they just had to basically buy a company to replace their technology or augment it, you know, whatever they say. But I mean it's like why did they have to make this huge acquisition

for something that is seems obviously duplicative to what they already have if they have such a mature offering.

Um yeah I mean I I I find that uh bizarre that they could possibly right here.

And I'm going to ask if you're listening uh let us know what you think. Let us know what you think but also if you want us to give a shot at creating our own magic quadrant. uh of of what we think the industry looks like today. Uh we'd be happy to give that a try if if you're interested in that. Um [snorts] yeah,

Diego's happy to give it a try.

I I'll be working hard. [laughter] I I'm sure this is no easy task. So, so no, you know, we love Gartner. We're going to be continue looking out for these insights. I think it's it really makes us think a lot about uh where things are today. So, uh with that being said, uh thanks for listening today. And if you have any other stories you want us to touch on, if you want uh us to interview anybody in particular, we'd be happy to hear your suggestions. Um and whether you're listening or watching, uh we'd appreciate your like, your comment, and your uh your thoughts. So, thanks.

Yeah. Thank you.