Third Bridge provides integrated research solutions for the world’s top investors and business leaders. Our expanding suites of products can help you drive change, putting crucial insights, value chain maps and experts at your fingertips.
The following transcript is generated from Third Bridge Forum, the biggest archive of high-quality expert interviews in the world, providing investors with critical knowledge to make smarter and faster decisions.
Submit your details to gain access to our premium content. If you would like to discuss our coverage of companies and sectors and how we can help you move faster and with greater conviction, please contact us today.
Specialist
Former Senior Executive at Intercom
Agenda
- Cost optimisation expectations around generative AI
- Generative AI-fuelled chatbots – commercialisation timeline and pricing expectations
- Generative AI performance trends and development
- Companies being impacted by generative AI and expected short-to-medium-term winners and losers
Questions
1.
How are you thinking about AI today? Could you identify 2-3 key trends that you think investors should be paying attention to when it comes to AI and its usage, development and trajectory?
2.
On the chatbot and customer service side specifically, over the last six months since our previous Forum Interview [see Conversation AI – Enterprise Chatbots & Customer Service Software – Emerging Generative AI Solutions & Potential Impacts – 6 March 2023], could you discuss what the progression has been in terms of utilising generative AI technologies within solutions as opposed to the discriminative AI that have historically been relied on?
3.
What hurdle do you think we have to get over to allow a player such as Salesforce or LivePerson to fully release a generative AI chatbot that might take us from 30% of conversations being answered with AI to 60-70%? We discussed in the aforementioned previous Forum Interview that this is what we’re working towards.
4.
In the next 12-18 months, even if we don’t reach the 60-80% we discussed, do you think we’ll start to see some of the cost optimisation that people envisioned when they first saw this generative AI, in terms of, “We can lay off 10% of our staff and save money by using technology”? Or do you not think we’re going to see actions such as that in the short term?
5.
One question that a lot of people have and that reports are starting to speculate about is what the pricing models and changes are going to be. Obviously, if you can lay off 10-20% of your staff, that’s 10-20% fewer seats that a chatbot software is going to support. I know Five9 made a change where its AI agent roll-out is going to be around 2-3x the price. Generally, how do you expect software pricing to trend as these cost optimisations begin to be recognised?
6.
When we say that if a company can save 10%, it might spend 20-30% of that savings on software, how do you think that could impact the revenue that chatbot and other support software is able to generate as generative AI solutions are brought to market?
7.
Is what we’re discussing a scenario where we might see one step back to take two steps forwards? Or, as people start to lay off customer service agents, will the pricing already be in place where it’s immediately starting to see revenue tailwinds?
8.
I keep pointing to LivePerson because it’s a core vendor that’s also a public company. Would you say it’s going to be another year or so before the company starts to include generative AI tailwinds in its revenue forecast?
9.
You mentioned the voice side of the industry. Obviously, contact centres remain a huge expense for a lot of companies. What level of impact do you think generative AI is going to have in replacing humans on the contact centre side, not just chat support agents? LivePerson talked a lot about the tailwinds there and the development of its solution. What are your expectations for that side of this industry?
10.
When we say we would love to get from 30% of AI answering questions today to 60-80%, do you think voice could be encompassed in that? Could we get to 80% of all inquiries – whether it’s e-mail, chat or voice – being answered, or is it going to take a much longer period of time to get rid of more human support call agents?
11.
In the last six months since our previous Forum Interview, in terms of the development of generative AI and finding ways to commercialise it and fuse it into products, would you say that’s gone about how you expected in terms of the speed of development? Or have there been more significant hurdles to commercialising it than we might have thought 6-8 months ago?
12.
Last time we spoke, one of the key challenges with commercialising any generative AI solution was the hallucination problem and how enterprises are just unwilling, especially on the support side, to give incorrect answers. Is it reasonable to expect that we’re going to have largely solved the hallucination challenge over the next 12-18 months, as broader generative AI solutions start to be rolled out? Or is it going to be much longer than that before enterprises are trusting gen AI to communicate with their customers and handle other inquiries?
13.
You mentioned the recent announcement by OpenAI and ChatGPT about GPT Enterprise. Could you share your reaction to that and your expectations in terms of the adoption and potential monetisation of this product roll-out?
14.
In terms of use cases, has what you’re discussing unlocked anything, or is it more so that organisations can let their employees use it and type in confidential data without the risk of it being distributed now?
15.
We spoke about the concept of internal LLMs [large language models] for companies. Is this the type of solution that’s going to plug into CRM, ERP [enterprise resource planning], Slack channels and so on, and just not be distributed elsewhere, when you say it’s not going to train based on data? Or will it have access to that information? Could you help us understand what an internal LLM does to then start thinking about what could be built on top of that?
16.
A specialist in a previous Forum Interview talked about how he doesn’t see a SaaS-like tool being used for internal LLMs, but rather companies themselves are building agents or capabilities on top of open-source models to avoid giving a third party such as OpenAI access to all their internal systems and data. What’s your take there, and what do you think the development of these internal tools and query bots is ultimately going to be built on top of?
17.
Ultimately, what type of LLMs do you think are going to win in the medium-to-long term? Is it the large ones from OpenAI or Google or, as the leaked Google memo said, do you think it’s going to be open-source and that OpenAI will have to find ways to offer more of a SaaS-like solution rather than the really broad LLM to generate revenue in the long term?
18.
What do you see as a future business opportunity for the LLM leaders? It feels as if there’s going to be so much built on top of them. Are these just going to become billion-and-billion-dollar business segments, as AWS and Azure have become?
19.
We’ve seen developments verticalise LLMs, and we’re seeing investments in general ones that are trying to compete with players such as OpenAI and ChatGPT. Do you think that’s a smart place to be investing? In the long term, do you think there’s going to be a whole handful of different LLMs that are used and that make a ton of money, or is it going to be more of the 3-7 that went out and everyone else will just not be able to compete as these models continue to train based on their enormous usage?
20.
Shifting over to the spending side, I think you said 70% of companies plan on spending on AI over the next 12 months. Where do you think most of that spend is going to be directed? We’ve seen enterprise software companies such as Salesforce or Microsoft come out with different features. They’ve started rolling out pricing. Do you think that’s where most of the spending will be directed? Is it going to be on services hiring a C3, Palantir or larger consultancy to come in and build agents on top of these internal models? What do you think that dynamic is going to look like and how might that trend?
21.
In terms of the timeline, I spoke to someone anecdotally who said we’re seeing a ton of the trials and proposals, but we haven’t really seen people make an investment yet in a large-scale AI tech transformation or project. When do you think we’re going to start seeing an acceleration of that type of work? What companies might be best-positioned to benefit from that when it does begin?
22.
What type of projects do you think are going to be undertaken first? Is it just upgrading to Microsoft’s Copilot and spending USD 30 more per month, or Salesforce’s Genius, or is it going to be more customised internal LLMs? What do you think will be the priorities once decisions around budget allocation are made?
23.
Do you think the offering from OpenAI will be something most companies are utilising within a six-month period?
24.
Do you think the larger tech companies are going to have an unrivalled advantage to leverage AI to drive more revenue, at least in the tech sector, or do you think the winners could be the smaller companies? I think you mentioned Typeface. What about companies that many people haven’t heard of yet? Who do you think are going to be the short-to-medium-term winners of this AI spending revolution?
25.
In our previous Interview, you noted HubSpot as one of the early leaders in terms of utilising AI and were impressed with what this company was doing. Are there any other companies that have rolled out products you’ve been impressed with in terms of utilisation of AI, or companies that you think have a particularly interesting opportunity to leverage it moving forwards?
26.
In terms of adjacencies, you mentioned six months ago that Nvidia was really well-positioned. I think we all know how the company has performed over the last six months. Is there anyone else in adjacent spaces – whether security, compute, data or chips – that you think is positioned to feel tailwinds over the next 6-18 months related to AI?
27.
Ultimately, on the chatbot side of the industry, what are your thoughts on the ultimate growth trajectory? Thinking about incumbents such as Salesforce, LivePerson and Ada, when might we expect to see some type of growth acceleration stemming from this AI, if ever?
28.
In our previous Interview, we mentioned how it was feeling a bit like a gold rush for AI investing and how there are a million AI companies with half-baked ideas. You said you have to identify who’s just a pretty face on top of ChatGPT. With a VC or PE audience in mind, what would be your criteria right now as an investor selecting AI start-ups and projects to be investing in?
29.
We’ve talked about a lot of opportunities so far. You floated one risk – potential regulation. Do you have any additional comments around what to monitor here or how relevant that risk is in terms of monetisation opportunities and potentially losing these revenue streams as we start to see the government get more involved in regulating this industry?
30.
Is there anything we haven’t covered or points you want to put out there that we haven’t done justice to so far around generative AI?
Gain access to Premium Content
Submit your details to access up to 5 Forum Transcripts or to request a complimentary 48 hour week trial
The information, material and content contained in this transcript (“Content”) is for information purposes only and does not constitute advice of any type or a trade recommendation and should not form the basis of any investment decision.This transcript has been edited by Third Bridge for ease of reading. Third Bridge Group Limited and its affiliates (together “Third Bridge”) make no representation and accept no liability for the Contentor for any errors, omissions or inaccuracies in respect of it. The views of the specialist expressed in the Content are those of the specialist and they are not endorsed by, nor do they represent the opinion of, Third Bridge. Third Bridge reserves all copyright, intellectual and other property rights in the Content. Any modification, reformatting, copying, displaying, distributing, transmitting, publishing, licensing, creating derivative works from, transferring or selling any Content is strictly prohibited
