Speakers discuss AI shift to edge devices, robotics commercialization, and data center regulatory risks in 2026
Summary
Host Jeremy Maggs interviews David Smith and Zane Bezuidenhout about the state of AI in 2026. Smith argues that useful intelligence has become cheap enough for ubiquitous consumer agents, while Bezuidenhout highlights the hardware shift toward edge AI via NVIDIA and AMD partnerships. The discussion covers the commercialization of robotics, with Smith citing Symbotic's success in enterprise automation. They also address Alphabet's market reaction to slowing search growth despite strong cloud performance, and debate the political and environmental impacts of data center expansion.
Key takeaways
- 1 Smith argues Google is a 'full-stack' AI winner with its own chips, cloud, and models, pointing out that Google reported its fastest search growth in five years, largely driven by AI mode queries.
Unidentified speaker (speaker_1) · 7:16
Episode sections
- 0:01 – 2:06
Introduction and 2026 Tech Definition
Host Jeremy Maggs introduces guests David Smith and Zane Bezuidenhout to discuss AI developments in 2026. David Smith identifies the year as when useful intelligence became cheap enough for ubiquitous use, enabling consumer agents to perform tasks like booking restaurants.
- 2:06 – 3:24
Edge AI and Hardware Shifts
Zane Bezuidenhout highlights NVIDIA's partnership with Microsoft for AI laptops and AMD's Ryzen capabilities as key developments. He notes this shifts AI from data centers to end-user devices, reinforcing agentic AI adoption and paving the way for physical AI like robotics.
- 3:24 – 5:43
Robotics Milestones and Commercialization
David Smith argues robotics is already scalable in three buckets: autonomous vehicles (citing Tesla's growth), enterprise robotics (citing Symbotic's automation for Walmart), and humanoid robots. He notes humanoid robots are still 5 years away from full autonomy due to complexity.
- 5:43 – 7:58
Alphabet's Market Reaction and AI Strategy
Smith discusses Alphabet's market value loss due to slowing search growth (12% vs 16%) and high cloud spending. He argues Google is a full-stack winner, citing record search volume in AI mode, but monetization lags behind growth.
- 7:58 – 9:57
Political Risks and Data Center Regulation
Smith identifies regulatory delays in data center rollout as the primary political risk for tech investors. Bezuidenhout adds that regulation, if done correctly, could enhance long-term adoption by ensuring safety, similar to motor vehicle rules.
- 9:57 – 12:24
Data Center Community Impact and NIMBYism
Smith addresses NIMBYism regarding data centers, citing noise and resource concerns. He suggests direct community investment and grid balancing could cut local electricity prices by 10%. Bezuidenhout clarifies data center water use is low compared to US golf courses due to recycling.
- 12:24 – 17:39
Workforce Impact and Data Preparation
Bezuidenhout discusses the shift from execution to judgment in jobs and the need for data cleaning. He mentions Snowflake and private competitor Databricks as key enablers for preparing corporate data for AI agents.
- 17:39 – 31:58
AI Governance and the 'Harness Layer'
Smith emphasizes the need for human-in-the-loop oversight for AI agents. Bezuidenhout introduces the 'harness layer' concept, involving cybersecurity, observability, and orchestration to prevent agent misbehavior, citing Anthropic's Mythos model launch.
Assets discussed
| Asset | Kind | Stance | Context |
|---|---|---|---|
| Anthropic | private company | neutral | So, David, back to you and Anthropic's preliminary IPO prospectus gives us a closer look at the business behind Claude. |
| OpenAI | private company | neutral | of we've been talking about anthropic or it's a competitor OpenAI or a few others in that space. |
“Stance” is how the speakers talked about the asset, not a recommendation.
Claims worth verifying
2026 was the year useful intelligence became cheap enough for ubiquitous use, enabling agents like MetaMuse and Instinct to perform tasks like booking restaurants.
Unidentified speaker (speaker_1) ·1:15
How to verify: Check consumer app availability and feature sets for MetaMuse and Instinct in 2026; review tech industry reports on AI inference cost reductions in 2026. For recent events, verify 2026-08-04 through 2026-11-04. Use another period only when the speaker explicitly names it.
NVIDIA announced a partnership with Microsoft around AI laptops, and AMD announced similar Ryzen capabilities for personal computing chips.
Unidentified speaker (speaker_2) ·2:06
How to verify: Search NVIDIA and Microsoft press releases for AI laptop partnerships in 2026; check AMD announcements for Ryzen AI capabilities in 2026. For recent events, verify 2026-08-04 through 2026-11-04. Use another period only when the speaker explicitly names it.
Primary sourceSymbotic's revenue grew ten times over a period of four to five years by automating warehouses for Walmart.
Unidentified speaker (speaker_1) ·4:04
How to verify: Review Symbotic (Symb) SEC filings (10-K/10-Q) for revenue growth rates and customer disclosures regarding Walmart. For recent events, verify 2026-08-04 through 2026-11-04. Use another period only when the speaker explicitly names it.
Primary sourceAlphabet's Google search business growth decelerated from 16% to 12% year-on-year.
Unidentified speaker (speaker_1) ·6:03
How to verify: Review Alphabet (GOOGL) Q3/Q4 2026 earnings call transcripts or SEC filings (8-K/10-K) for search revenue growth metrics. For recent events, verify 2026-08-04 through 2026-11-04. Use another period only when the speaker explicitly names it.
Primary sourceAlphabet's Google Cloud business grew 100% year-on-year.
Unidentified speaker (speaker_1) ·6:23
How to verify: Review Alphabet (GOOGL) earnings reports for Google Cloud segment revenue growth. For recent events, verify 2026-08-04 through 2026-11-04. Use another period only when the speaker explicitly names it.
Google reported its fastest search growth in more than five years, with many searches in AI mode.
Unidentified speaker (speaker_1) ·7:16
How to verify: Review Alphabet (GOOGL) earnings call transcripts for management commentary on search volume and AI mode usage. For recent events, verify 2026-08-04 through 2026-11-04. Use another period only when the speaker explicitly names it.
Using data centers to balance the grid in a regional area can cut electricity prices by 10%.
Unidentified speaker (speaker_1) ·10:48
How to verify: Search for studies or reports from energy regulators or data center operators on grid balancing benefits and price impacts in 2026. For recent events, verify 2026-08-04 through 2026-11-04. Use another period only when the speaker explicitly names it.
Data center water consumption is significantly lower than golf course water consumption in the United States due to recycling.
Unidentified speaker (speaker_2) ·11:39
How to verify: Review industry reports from tech companies or water usage studies comparing data center cooling water use to US golf course consumption. For recent events, verify 2026-08-04 through 2026-11-04. Use another period only when the speaker explicitly names it.
Anthropic reached an annual revenue run rate of approximately $70 billion.
Unidentified speaker (speaker_1) ·20:40
How to verify: Review Anthropic's preliminary IPO prospectus (S-1) filed with the SEC for revenue figures. For recent events, verify 2026-08-04 through 2026-11-04. Use another period only when the speaker explicitly names it.
Primary sourceAnthropic has 11,900 clients spending an annualized amount of $100,000 or more.
Unidentified speaker (speaker_1) ·21:00
How to verify: Review Anthropic's preliminary IPO prospectus (S-1) for customer concentration and spending data. For recent events, verify 2026-08-04 through 2026-11-04. Use another period only when the speaker explicitly names it.
Do your own research
- Verify Alphabet's Q3 2026 earnings report for search growth rates (12% vs 16%) and Google Cloud YoY growth figures.
- Check NVIDIA and Microsoft's official press releases regarding AI laptop partnerships and hardware specifications.
- Review Symbotic's latest 10-K or 10-Q filings for revenue growth metrics and Walmart contract details.
- Assess regulatory filings for data center permits in key regions to evaluate NIMBY and zoning delays.
- Compare data center water usage statistics from industry reports against US golf course water consumption data.
- Read Anthropic's technical documentation on the 'Mythos' model and 'harness layer' cybersecurity features.
- Investigate Shopify's public statements or earnings calls regarding Toby Lutke's comments on AI responsibility and human-in-the-loop requirements.
Full transcriptexpand
- 0:01
- Unidentified speaker (UNKNOWN): As
- 0:02
- Unidentified speaker (speaker_0): 2026 draws to a close, what have we learnt about artificial intelligence's potential, its costs, and the businesses being built around it? When we last spoke to today's guests, they argued that artificial intelligence could drive decades of growth. They also pointed to a world in which digital employees would start taking on
- 0:22
- Unidentified speaker (speaker_0): more of our work. Now it's time to revisit those expectations. Which developments this year have strengthened their conviction and which have given them reason to reconsider? How do they distinguish a promising technology from a business that can turn that promise into lasting profits? With Americans just having voted in the U.S.
- 0:41
- Unidentified speaker (speaker_0): midterms, we'll ask what the political backdrop could mean for the AI investment boom, whether we're in a bubble, and what to watch out for in 2027 and the next
- 0:50
- Unidentified speaker (UNKNOWN): year.
- 0:51
- Unidentified speaker (speaker_0): five years. Hello, I'm Jeremy Mags. You're listening to No Ordinary Wednesday. And joining me from London are Investec Global Equity Analysts David Smith and Zayn Bazadenhoot. Gentlemen, welcome back to the show. David, first question to you then, and if you had to choose one event or development that defined
- 1:10
- Unidentified speaker (speaker_0): technology in 2026, what might it be and why?
- 1:15
- Unidentified speaker (speaker_1): Jeremy, Zayn and I were debating this before we came on your podcast. And I'm probably gonna have to stick with 2026 was the year when useful intelligence became cheap enough that we could use it almost ubiquitously. It is the year where it became cheap enough to get
- 1:35
- Unidentified speaker (speaker_1): agents to start doing things for consumers. So we've had a couple of agents out MetaMuse, there's another one called Instinct, which I use. And while these companies still aren't making money off it, they are able to actually afford to have agents that will start doing tasks for you, like booking restaurants and filling out diary requests, which is pretty cool for me because it
- 1:55
- Unidentified speaker (speaker_1): means I can finally actually be decent at admin.
- 1:57
- Unidentified speaker (speaker_0): So Zayn, let me reverse engineer that question then. Was there something that took place this year then? that didn't get as much recognition and maybe should have?
- 2:06
- Unidentified speaker (speaker_2): Yes, uh thank you, Jeremy. As Dave mentioned, we were chatting, and I think maybe it was I think around NVIDIA's big technology event in Taipei and Taiwan, and they announced a partnership with Microsoft around AI laptops, basically. And their competitors, AMD, have announced a similar type of product in
- 2:26
- Unidentified speaker (speaker_2): terms of their personal computing chip Ryzen capabilities. But basically, this means that consumers like you and I can run maybe last year's frontier models on our laptops. And I think this is maybe an important development in that it now moves AI capabilities away from data centers and the
- 2:46
- Unidentified speaker (speaker_2): cloud closer to us, the end users. And that is reinforcing of the adoption and proliferation of what Dave mentioned around being agentic AI. And it has consequences for, you know, and I'm sure we'll discuss it, the next step being physical AI, whether that's robotic or autonomous vehicles.
- 3:04
- Unidentified speaker (speaker_0): Alright, David, when we spoke in July, if memory serves, you highlighted robotics as a longer term opportunity. So a couple of months on, what specific commercial market? milestone or milestones would persuade you that things are moving beyond promising demonstrations into something that is a little more
- 3:23
- Unidentified speaker (speaker_0): scalable?
- 3:24
- Unidentified speaker (speaker_1): Jeremy, I actually think we're there. Now we're still very early days, and that that context is important. And when we think about robotics, there's probably three key buckets. The bucket one would be autonomous vehicles. Think about the full self self-drive drive that Tesla. has. That's sort of grown three to four times in four years at the current trajectory. I would be very
- 3:44
- Unidentified speaker (speaker_1): surprised if most people who buy a new car in five years don't have access to full self-drive in more than half of vehicles. So that's happening already. The second bucket is what you sort of classify as enterprise robotics. Think about a really fancy manufacturing site or a distribution center where you'd have robots that would be doing the vast majority of the
- 4:04
- Unidentified speaker (speaker_1): work. Musk famously has all these pictures and videos of what happens in his factories, but you have the likes of Symbotic in the US who fully automates or close to fully automates warehouses for the likes of Walmart. And that's growing really nicely. I mean it's up sort of ten times on revenue in a space of four, five years. And we think that traction of that's going
- 4:23
- Unidentified speaker (speaker_1): to continue to accelerate. The last one is one that most people think of, though, and that's humanoid robots. So I'm guessing most people listening to the podcast have seen context or videos where you see robots having robot Olympics or being in people's houses where you'd be vacuuming. And that's starting to happen. The problem is
- 4:43
- Unidentified speaker (speaker_1): that the complexity of the fringe cases. Think about how when you pick up a glass on a table to clean underneath it, the pressure you use on that. glass could break it. So there's a whole bunch of software that needs to be happening in the background so these robots can know when to do things. They are actually interestingly the companies right now in San Francisco that'll come to your house and clean it for thirty dollars an
- 5:03
- Unidentified speaker (speaker_1): hour with a robot. It's very early, but the problem is they have to have somebody in the background actually controlling the robot in certain instances. So I suspect that's probably at least five years out. But I'd be very surprised if we have a chat in five years' time where you don't have at least one friend that has a fully autonomous robot in their house.
- 5:23
- Unidentified speaker (speaker_0): All right. David, Alphabet suffered a sharp loss in market value following its latest results, despite strong growth in Google Cloud. Let me ask you this: what worried investors more? Was it the cost of its AI ambitions or the threat that AI poses to the search advertising business that helps fund them?
- 5:43
- Unidentified speaker (speaker_1): Jeremy, um you're talking to a tech investor, so the last week or two has been particularly painful. I feel like I've got a couple of scars in my back from the gyrations in the market. So it definitely was Alphabet, but it's happened to more than Alphabet. It was one of the worst weeks for tech that we've had this year. But Alphabet in particular had two issues. One is that the
- 6:03
- Unidentified speaker (speaker_1): Google search business, the one that you traditionally type into a browser, that had one of the slowest growth rates we've seen in a while. It deaccelerated from 16% the prior quarter to 12, and it suddenly got investors very nervous that AI is going to eat into their search business. You won't have to go into Google anymore to ask questions because
- 6:23
- Unidentified speaker (speaker_1): you can just ask your AI to do it instead. The other bit is they are spending a lot of money on their cloud business. That grew amazingly. It grew nawful 100% year on year. But people don't tend to like to when they're a little bit nervous with the market, they don't like to invest in businesses that are free cash flow negative. And they're probably going to remain that
- 6:43
- Unidentified speaker (speaker_1): way for at least a year or so.
- 6:45
- Unidentified speaker (speaker_0): So what would you need to see then in the next set of results to judge whether AI is strengthening Google's business or eroding its earnings power?
- 6:56
- Unidentified speaker (UNKNOWN): So
- 6:56
- Unidentified speaker (speaker_1): Zane and I spend a lot of time trying to answer that exact question. We are fairly comfortable that Google's going to be one of the major winners in AI in the next five years. They've got this full-tech stack approach where they have their own chips, they have their own cloud, they have a whole bunch of users they can drive it to, they've got the models. They basically one of their only the only full stack approaches
- 7:16
- Unidentified speaker (speaker_1): towards AI. And we think that that's coming through in a number of ways. So what's really interesting is you talked, well, I spoke about the Google search revenue growth. But what management highlights on their recent call is they actually had the fastest growth in searches they've had in more than five years. A lot of those searches were in AI mode. So we are firm believers that
- 7:36
- Unidentified speaker (speaker_1): they're going to be a winner, but it's just going to take a little time. before they pull the monetization lever.
- 7:41
- Unidentified speaker (speaker_0): And David, Americans voting in the midterms that just voted, what would you be watching as the outcome consequences become clearer? And how could the balance of power in Congress then affect AI investment, maybe through spending decisions or regulation or even borrowing costs?
- 7:58
- Unidentified speaker (speaker_1): Jamie, Jeremy, so I think the number one concern that tech investors have is that you have some sort of wobble in terms of supply of data centers coming online. So that would have a ripple effect throughout the supply chain, the chip makers, the model makers, if you just can't produce enough units of intelligence or tokens because
- 8:18
- Unidentified speaker (speaker_1): you just don't have enough data centers coming online. So the worry always is that you have some sort of regulatory overview from whichever body in whichever country that slows down the rollout of data centers. So that's probably front of mind for everyone. It tends not to be at state level, at a higher level in the US, it tends to be very specific to states or
- 8:38
- Unidentified speaker (speaker_1): counties, rather. So it would be a case of do you see a change in politicians in charge slowing down the rollout royal out of data centers and making it more difficult?
- 8:48
- Unidentified speaker (speaker_0): So Zane, Zayn, just a quick follow follow-up up then to what David has been saying. how much difference can Washington make to an investment boom being driven by private companies?
- 8:56
- Unidentified speaker (speaker_2): I think not just specific to Washington, but any government. I think regulation is necessary to some degree, especially for nascent industries and particularly one of this nature where we're dealing in the currency of intelligence. You know, the impacts can potentially blunt or dampen short term. short-term growth. But if done correctly,
- 9:16
- Unidentified speaker (speaker_2): and that's caveated with a big if, if done correctly, then they could make the growth, although blunted in the short term, more durable and ultimately drive long term adoption to a technology. I always like to use the example when motor vehicles were invented, rules of the road weren't invented coincidentally. You
- 9:36
- Unidentified speaker (speaker_2): know, that had to be regulated for everyone's safety. And when it is safe to use, that drives further adoption. So if done correctly, regulation could actually be an enabler of adoption and proliferation of the technology going forward.
- 9:57
- Unidentified speaker (speaker_0): Data centers and the prospect of higher electricity bills. So here's my question: how should the cost of expansion? expanding local infrastructure be shared between tech companies and communities?
- 10:08
- Unidentified speaker (speaker_1): There's a massive drive of NIMBYism right now. For those who don't know, the acronym NIMBYism stands for not in my backyard. You spot on. There's a resistance to it. And there are a number of reasons for it. One, if you live very close to a data center, they tend to have huge amounts of generators in the background, and those generators make a lot of noise. So there's no doubt there's some of that.
- 10:28
- Unidentified speaker (speaker_1): But there's also a lot of concern around the usage of water or electricity. The data shows it's actually not as bad as often feared, and often happens it's blown up in the press. But to get around those issues, what I think should absolutely be happening, is that you have direct investment from the data center companies into the communities to help with
- 10:48
- Unidentified speaker (speaker_1): infrastructure and services around the area. But they there's actually some really cool data around things like if you use the data center to balance the grid in a regional area, you can actually cut electricity prices by 10%. It does, however, require coordination with the state bodies. And that's often the trick. But to get around these issues where people do
- 11:07
- Unidentified speaker (speaker_1): push back on data centers and areas, I think it's almost going to become a prerequisite. Mm.
- 11:12
- Unidentified speaker (UNKNOWN): So
- 11:12
- Unidentified speaker (speaker_0): Zayn, Zay, let me ask you this then what should companies be expected to demonstrate about their water use before expanding into water stressed areas?
- 11:21
- Unidentified speaker (speaker_2): Been quite a little bit maybe of uh misinformation around this. Yes, data centers do use uh water, but a lot of that is recycled within their systems. So they're not big consumers of net new water resources. Yes, new water resources are needed, but also because of
- 11:38
- Unidentified speaker (UNKNOWN): the water.
- 11:39
- Unidentified speaker (speaker_2): that ability to recycle water mainly for cooling within the system, you know, their their consumption isn't large. And I've seen studies where they put the water use in comparison just to the golf courses in the United States. And the data center water consumption pales to the golf course water consumption in the United States. So in
- 11:58
- Unidentified speaker (speaker_2): terms of it it seems to have been maybe a political sticking point, but reality is far different from what the narrative might suggest.
- 12:06
- Unidentified speaker (speaker_0): And Zayn, if AI takes over the junior tasks through which people traditionally learnt, how can employers, I wonder, develop the next generation of specialists without the expertise to challenge its answers rather than simply rely on them?
- 12:24
- Unidentified speaker (speaker_2): Yeah, it's a tough one. We do see it. We do hear it. anecdotally, even in our industry, around the number of interns that are absorbed by corporations is lessening. That being said, I think like with any technology shift, and I remember quite clearly the world that my parents ran, their business in South Africa versus the one that I
- 12:44
- Unidentified speaker (speaker_2): walked into was chalk and cheese. I mean, an electronic typewriter was a modern invention in their time and use of facts. I think in terms of the doing bit, the workflow bit, that maybe gets taken away, but you still need an understanding of workflow processes. And I think the experience side shifts
- 13:04
- Unidentified speaker (speaker_2): more towards judgment, management, and delegation. And I think it's also just a bigger The question, not only around what they learn in the early stages of their careers, but also what the education system brings to young people for the age of AI. And you know, probably a lot of reinvention that is required in that regard.
- 13:23
- Unidentified speaker (speaker_0): You're listening to No Ordinary Wednesday on Investec Focus Radio with David Smith and Zayn Bazadenhote. Now after the break, we're going to explore how businesses can judge AI's reliability, examine the economics of AI companies, and look ahead to 2027 and beyond. First, discover Investec Wealth and Investments Global Leaders
- 13:43
- Unidentified speaker (speaker_0): Portfolio. Managed by an experienced team in South Africa and the UK, the portfolio invests in 30 to 50 high-conviction global companies with enduring competitive advantages, taking a long-term approach to capital growth. To find out more, visit Investec.com.
- 14:00
- Unidentified speaker (UNKNOWN): The
- 14:00
- Unidentified speaker (speaker_0): Global Leaders Portfolio invests. exclusively in equities and carries a high degree of risk. The value of investments can go down as well as up, and investors may get back less than they invested. Past performance is not a reliable indicator of future results. Now let's get back to the conversation. All right, Zayn, back to you and how much of
- 14:20
- Unidentified speaker (speaker_0): the difficulty then in adopting AI comes down to a company's own data being fragmented. outdated, or poorly organized. And I wonder whether businesses are underestimating that work.
- 14:33
- Unidentified speaker (speaker_2): I think we are in this period of shifting from, at least in a corporate sense, and I know we are as a business, shifting from experimentation to implementation. And what that does highlight is what needs to be done in terms of going back to, you know, Dave's first comment around AI being useful. And that's a gentic AI. It's technically only
- 14:53
- Unidentified speaker (speaker_2): as useful as the data it has to work with and the understanding and the context of the business that it's operating within. There are some companies that assist with this on the public side. There's a company called Snowflake that helps corporations get their data ready. There's a private company called Databricks that you know competes with Snowflake. I mean, if you think about
- 15:13
- Unidentified speaker (speaker_2): you know data, if you just think about a spreadsheet for that matter, and you have today's date in one cell, but then you have it formatted in a whole bunch of different ways, maybe with a comma in there that makes it unreadable. It really is scrubbing the data and making it interpretable by your artificial intelligence agents and the models that
- 15:33
- Unidentified speaker (speaker_2): you put on top of it. So this is a very necessary part for corporations to make maximize the usefulness of AI that when they're integrated with their workflows as an organization.
- 15:45
- Unidentified speaker (speaker_0): All right. Zayn, let me ask you about operational context that you've just referred to. How then should businesses test AI systems that can give a convincing answer even when they're wrong? Are industries benchmarks then than measuring what matters in everyday business use, I wonder?
- 16:01
- Unidentified speaker (speaker_2): I think this is where subjectivity comes into it. So yes, models get measured by, you know, scientific scores and intelligence scores that are there as a standardized measure of capability. But really what, you know, consumers in their personal capacity or what corporations in an enterprise capacity want is usefulness. They want
- 16:21
- Unidentified speaker (speaker_2): outcomes. And they want the most correct outcome for a given scenario. And you know quite frankly, you know, in a human environment, it's not a case of optimizing, but more sometimes a case of satisfacing. So it's about finding a good enough solution to get the outcome that is most desirable. And I think how
- 16:41
- Unidentified speaker (speaker_2): businesses go about this is yes, there has to be strategic leadership and directions set in place, but ultimately the context of these workflows are by the users. So people like Dave and I, when it comes to researching companies, we know best how to measure the outcome that we want. And it might not be the perfect outcome. And
- 17:00
- Unidentified speaker (speaker_2): through lots of iterations, it will improve. So I think it's measuring AI around its usefulness in the eyes of the end user, which is us as employees in the organization.
- 17:11
- Unidentified speaker (speaker_0): Now, XMTP Labs chief executive should. Shane Mack, describing how an AI agent he used for personal and professional tasks, posted his bank account information in a company Slack channel. He said the agent appeared to be doing what it thought it should do. So, David, how should
- 17:31
- Unidentified speaker (speaker_0): businesses decide then what an AI agent is allowed to access and to do and when it must stop
- 17:38
- Unidentified speaker (UNKNOWN): the process?
- 17:39
- Unidentified speaker (speaker_0): and ask a human for approval. I sense this is a very important question.
- 17:44
- Unidentified speaker (speaker_1): Jeremy, you are a hundred percent right. So I would first make the argument that you should absolutely be separating your personal life from your work life, whether it's on an agentic level or otherwise. So I feel like that's sort of the what we all learned around the internet trying to separate the two lives and not use the same devices. or or the same systems
- 18:04
- Unidentified speaker (speaker_1): applies here as well. But Toby Lee Licker, who's the CEO of Shopify, made a really interesting comment. And his view is that AI can do amazing things and will continue to do amazing things. What it can't do is take responsibility. So in my mind, we have moments where an a output is created or
- 18:24
- Unidentified speaker (speaker_1): something is used that is going to impact. a decision, a workflow that is not automated or normal. You absolutely need a human in the loop to check. And I think that's probably how it plays out is that you're going to have all these systems which are automated, but you're going to have to continually have them checked, one, by other agents, but two, ultimately by a
- 18:44
- Unidentified speaker (speaker_1): human to make sure that the errors that come through, just like you wouldn't trust a junior to publish a company's view. to clients without some sort of oversight, the same is going to be true for for agents.
- 18:55
- Unidentified speaker (speaker_2): Yeah, if I may chime in, what Dade said is spot on. And what we are seeing, particularly on the certain areas within the software stack and where not all software companies are threatened by AI, is I think there's a growing appreciation, particularly at an enterprise level, you know, the industry's industry is calling this the harness layer. And this is
- 19:15
- Unidentified speaker (speaker_2): really around cybersecurity. And one can see what the share prices of cybersecurity companies have done since Anthropic launched their Mythos model and also project glass wing off the back of that to select users to patchwh
- 19:28
- Unidentified speaker (UNKNOWN): where
- 19:28
- Unidentified speaker (speaker_2): it's revealed certain cracks in their security architecture. There's also observability. So watching these agents as they move. And then there's an orchestration element. with governance also built in, and that's company policies. So it is almost wrapping these agents up in cottonwool to try and ensure that they
- 19:48
- Unidentified speaker (speaker_2): behave how they should behave. And ultimately, like David said, and then a human in the loop, depending on the use case and the workflow, just to make sure that none of that wrapping has fallen by the wayside and that the agent is doing something it shouldn't. So as you said, a very important question, and the industry's increasingly talking around this harness layer in a way
- 20:08
- Unidentified speaker (speaker_2): to protect users, companies, individuals from agents misbehaving in pursuit of achieving an outcome.
- 20:15
- Unidentified speaker (speaker_0): Now, before we discuss individual businesses, I just want to remind you this is analysis, not investment advice. Listeners should do their own research. So David, back to you and Anthropic's preliminary IPO prospectus gives us a closer look at the business behind Claude. Very interested to know what you've learned from it.
- 20:35
- Unidentified speaker (speaker_0): What did you find compelling? And anything there that might have given you pause.
- 20:40
- Unidentified speaker (speaker_1): It's one of the most exciting growth stories you're going to ever see as a potential investor. The sheer run rate at and how fast this company is has growing grown from having no revenue. in 2022 to being at an annual run rate of call it $70 odd billion dollars right now. Just phenomenal. The second bit, which is quite encouraging, is they have
- 21:00
- Unidentified speaker (speaker_1): $11,900 clients who spend an annualized amount with them of $100,000 or more. So the breadth of their customer base is really big. The bits that worry both Zane and I actually are more around the long term modes. Okay, if you're talking around defensibility of your business model
- 21:20
- Unidentified speaker (speaker_1): in the long run and how what quality you have to protect yourself. What we've seen the last couple of weeks, the likes of Google and Meta have come out with amazing frontier end models that can compete with the best models for open air aeroanthropic. anthropic. Never mind what's happening in China, and they're having increasingly having really good models. of their own. So I
- 21:40
- Unidentified speaker (speaker_1): have some concerns around the fact that the model layer actually just becomes fairly normalized in terms of what it can do relative to other models, that they all become amazing and it actually just becomes a commodity that keeps us up awake at night. Do you need to invest in the model layer or can you just choose other parts of the ecosystem?
- 21:58
- Unidentified speaker (UNKNOWN): So,
- 21:58
- Unidentified speaker (speaker_0): Zayn, let me widen the aperture slightly. The persistent question on investors' minds is are we in an AI bubble? And I'm wondering if the technology can live up to its promise while investors still overpay for the businesses building it.
- 22:14
- Unidentified speaker (speaker_2): Jeremy, yes, that's certainly been the question on markets' mind, particularly in the month of October when we've seen quite a big sell-off in this tech space. And I would say investors are not overpaying, especially on the public side. And there's maybe some bifurcation here. You know, there's been a lot of valuation momentum in private companies. And what we've seen in the likes
- 22:34
- Unidentified speaker (speaker_2): of we've been talking about anthropic or it's a competitor OpenAI or a few others in that space. We've seen their valuations ratchet up quite meaningfully. You know, in the public space, we've seen similar growth, but what we've observed is actually that growth has been led by earnings. So it's been earnings fundamental. earnings that have driven that growth appreciation in
- 22:54
- Unidentified speaker (speaker_2): these companies' valuations. And if you look at a you know the poster child of artificial intelligence being NVIDIA, you can buy it on a forward 14 times PE as we speak at the moment. You know, some of the companies in the memory space, you know, they're trading on a forward 4P PE of 2.9 times. So I think the market environment now
- 23:13
- Unidentified speaker (speaker_2): is that investment. investors have been overly skeptical. Well, I think the sentiment's gone the other way and fearing that the earnings power of these companies cannot persist for much longer. You know, based on our work that Dave and I have done, we've basically got NVIDIA going to the shape, its financials in terms of margins and returns back to where it was
- 23:33
- Unidentified speaker (speaker_2): pre-AI after 28. So almost the party's over after 28. And when Dave and I step back and we look 10 years out and we look at the technological shift, this is really a fundamental shift in how we as human beings do things with readily available intelligence at our
- 23:53
- Unidentified speaker (speaker_2): fingertips. And we think that as a driver, when we're looking back in 10 Ten years' time, I think the world would be demanding more compute power in the form of more intelligent tokens. than what we currently are. And that kind of gives us longer term confidence for where we are at the moment.
- 24:08
- Unidentified speaker (speaker_0): So, David, a quick follow up, what would tell you that spending on AI infrastructure is running ahead of likely demand? And if expectations fall or borrowing costs stay high, which businesses are most exposed in your opinion?
- 24:22
- Unidentified speaker (speaker_1): This cycle is quite different to previous cycles when you've had lots of catpects. capex. Think about what happened with the internet fiber rollout, or if you had with railway networks, you go back a lot further. Those bits of infrastructure, you have to build out a whole chain. So think about connecting New York to Chicago by rail network. So you have to build an entire rail network, which took a long
- 24:42
- Unidentified speaker (speaker_1): time, took a lot of cash before you even got to a stage where people could build a a carriage to run on the network. What's different now is that as soon as people turn on data centers, there's a massive backlog, those chips light up immediately. So there's no dark chips at the moment out there where you've got excess compute that
- 25:02
- Unidentified speaker (speaker_1): people don't want to use. In actual fact, if you look at the ratio of backlogs for the cloud players relative to their revenue, it's almost never been higher. So we're comfortable right now that there's a mismatch in demand and supply. One of the data points we track. it's not going to be a perfect predictor, but it's useful anyway, is the spot prices of renting a GPU. And what we see there
- 25:22
- Unidentified speaker (speaker_1): is the prices continue to trend upwards. In other words, the demand for those GPUs is our stripping supply. And those who actually have them online can charge more and more each month. So we do track that, but fundamentally we think we are quite somewhere away from having too much capacity. If anything, we think we will have an undersupply of capacity for way longer
- 25:42
- Unidentified speaker (speaker_1): than the market expects. But when we get to the point, because there will ultimately be a point where we have too much, the entire supply chain is going to come under pressure. The very high margins you're getting from the chip and memory players will definitely be pressured. But we think we are much further away than many people expect.
- 25:58
- Unidentified speaker (UNKNOWN): All
- 25:59
- Unidentified speaker (speaker_0): right, a question to both of you, and let's look ahead five years, twenty 2031, thirty one, if AI assistants become commonplace, what happens to businesses that currently depend on customers visiting their websites, comparing products, or dealing directly with their staff? And where could South African businesses find an opportunity in that change? David, do you want to
- 26:18
- Unidentified speaker (speaker_0): pick that up?
- 26:19
- Unidentified speaker (speaker_1): The starting point is you need to be fit for the new world. So I think that there's a fairly high chance it of what will happen is your agents as a company will talk to your customers' agents. So you need to prep for that because a lot of engagement will happen on that side. What you tend to find, and I I've tried this out with my own agent instinct, is that I still want a range of choice. So
- 26:39
- Unidentified speaker (speaker_1): I recently booked a flight to New York to visit a friend. And as part of that process, I said, please find me a bunch of flights. This is when I want to fly. This is the time I want to arrive by. And it gave me a variety of options and I chose. So you still will have that final point where you might have the person go and pull the trigger, or your product will be on a short list of where
- 26:59
- Unidentified speaker (speaker_1): you are. What that means is you probably need to have background checks to make sure that your products are competitive price-wise and that they are not left behind and that they are fit for purpose when an agent comes knocking on your website.
- 27:13
- Unidentified speaker (speaker_0): Zane, Zayn, anything to add to that?
- 27:15
- Unidentified speaker (speaker_2): Yeah, I think it's also an exciting time for businesses. I mean, we discuss this a lot internally to enhance your productivity. It's teaming up with the agents. You know, agents will become colleagues. The agents that Dave and I use in the workplace are there to assist us to enhance our productivity. So economically I think where you find
- 27:35
- Unidentified speaker (speaker_2): economies that have been maybe constrained by a certain labor force or how much the labor force is skilled or unskilled, you you're basically putting the world's intelligence in the hands of employees and they understand what their business is trying to achieve. And as Dave said, yes, there will be agent to agent business, but also for your customer at the end
- 27:55
- Unidentified speaker (speaker_2): of the day, is choice. I mean, we already live in that world where, you know, dealing with an airline, you're probably dealing with a chatbot or an automated call. center up until a point where maybe the query is a little bit technical or is, you know, unique or bespoke that requires human intervention. And I think that's how this will evolve and progress is kind of
- 28:15
- Unidentified speaker (speaker_2): the more remedial tasks will shift to a more automated level. And then with increasingly, need, you know, complex or bespoke scenarios shifting to humans and then AI capabilities. will adhere where it's supporting the human employees to enhance productivity for the businesses. So we do view this as a productivity
- 28:34
- Unidentified speaker (speaker_2): enhancement over the long term.
- 28:36
- Unidentified speaker (speaker_0): So Zayn, let me stay with you and in assessing the different layers of the AI value then, would you rather own the companies selling the picks and shovels, the companies building the AI models, or the businesses deploying AI?
- 28:51
- Unidentified speaker (speaker_2): I think where we are in the current market environment, I think it's all right to be in the picks and shovels still. We think there's durability to this build out. It's not going to be in a straight line and there are going to be bumps along the way. And we think the market is paying investors for that at the moment, where things have sold off in the month of October. In terms of how we, you know,
- 29:11
- Unidentified speaker (speaker_2): our lens, you know, as long-term investors for our clients.
- 29:15
- Unidentified speaker (speaker_2): I think we, you know, over the long term would look to the businesses that are ultimately deploying AI. And a lot of these are your established tech businesses, your cloud service providers, or some of them the hyperscalers. You know, these are companies that are building an asset, a long term asset in the form of these data centers, which basically they are renting out
- 29:35
- Unidentified speaker (speaker_2): intelligence to the world. They have existing MOTAS motes and they are using this technological shift to enhance their earnings power and those motes going forward. So picks and shovels, yes, for now. But in terms of durability, we do like the big names sitting at the center and being your cloud service providers.
- 29:51
- Unidentified speaker (speaker_0): All right, gentlemen, we're coming to the end of this fascinating conversation. And David, let me come to you first. Give us one prediction for tech in 2027 that we can come back and test. What do you think is going to change and how will we know it's happened?
- 30:07
- Unidentified speaker (speaker_1): I think the spread of consumer agents is gonna be one of the things that will surprise you. So MetaMuse and Instinct, the two apps I talked about. There's some access, you can't even get them outside of Japan and the US at the moment on a meta side. Instinct is only invite, but my suspicion is that more than half the people.
- 30:27
- Unidentified speaker (speaker_1): you know in a year will have access to an AI agent that will do a whole bunch of tasks for them.
- 30:32
- Unidentified speaker (speaker_0): And Zayn, your prediction and which widely anticipated breakthrough possibly do you think will still be waiting to happen, shall we say 12 months from now?
- 30:42
- Unidentified speaker (speaker_2): I think I'd be guessing, I mean, the world in terms of anticipated breakthroughs, there's been a lot of chatter outside of the AI world but loosely connected in the quantum realm and the effectiveness and the usefulness and the readiness primarily of that. From what we hear, it does feel like things have been getting closer, but also the
- 31:02
- Unidentified speaker (speaker_2): commentary around that people have been saying that for some time. But it really does feel in terms of quantum computing that we'll still be waiting, but there'll be some progress that's been made.
- 31:12
- Unidentified speaker (speaker_0): Gentlemen, our first conversation was fascinating, and let me tell you both, this was even more so. David Smith, Zayn DeZadenhote, thank you both, and we are going to keep those predictions for the next time we converse. Thank you for listening, and for more conversations on the forces shaping business and the economy, follow InvestEc Focus Radio
- 31:32
- Unidentified speaker (speaker_0): on Spotify, Apple Podcasts, or YouTube. Until next time, goodbye.
- 31:41
- Unidentified speaker (UNKNOWN): The
- 31:41
- Unidentified speaker (speaker_3): views expressed are those of the contributors at the time of publication and do not necessarily represent the views of the firm and should not be taken as advice or recommendations. Investec Limited and subsidiaries, authorized financial service providers, registered credit providers, and long-term insurer.
More from Investec Focus Radio SA
Macro Monday Ep140: Is the market pricing in too many US rate hikes?
Episode discussion: Holdsworth argues US markets overprice Fed hikes and China faces domestic demand slump
Speakers’ claims: Chris Holdsworth opens Macro Monday Ep140 by questioning whether the market is pricing in too many US Federal Reserve rate hikes. He points to persistent US inflation data but argues that base effects from oil prices will cause year-on-year inflation to decline, making further hikes unlikely after a December pause. On the global front, Holdsworth highlights a severe slowdown in Chinese domestic demand, citing weak credit growth, high youth unemployment, and falling vehicle and house prices. He contrasts this with a 35% surge in Chinese exports, warning that this imbalance may negatively impact global commodity prices.
Investec Minds Ep 5 | Nick Holland on decisions that reshaped Gold Fields
Episode discussion: Nick Holland on Gold Fields' safety culture, Sibanye unbundling, and long-term mining strategy
Speakers’ claims: Nick Holland, former CEO of Gold Fields, discusses the importance of safety culture, citing a four-day production halt after a fatality at South Deep. He explains the 2013 unbundling that created Sibanye Gold to separate South African assets from international operations. Holland advocates for a profit-focused mindset over volume, requiring a 15% free cash flow margin. He also addresses the challenges of deep-level mining in South Africa and positions gold as a hedge against geopolitical and financial instability.