Where We Might Be Going With Powerful AI

If you didn’t know where AI might be taking humanity over the next few years have a read below.

The Incredible Journey Humanity Is On With AI

In terms of pure intelligence⁴, it [powerful AI] is smarter than a Nobel Prize winner across most relevant fields – biology, programming, math, engineering, writing, etc. This means it can prove unsolved mathematical theorems, write extremely good novels, write difficult codebases from scratch, etc.

In addition to just being a “smart thing you talk to”, it has all the “interfaces” available to a human working virtually, including text, audio, video, mouse and keyboard control, and internet access. It can engage in any actions, communications, or remote operations enabled by this interface, including taking actions on the internet, taking or giving directions to humans, ordering materials, directing experiments, watching videos, making videos, and so on. It does all of these tasks with, again, a skill exceeding that of the most capable humans in the world.

It does not just passively answer questions; instead, it can be given tasks that take hours, days, or weeks to complete, and then goes off and does those tasks autonomously, in the way a smart employee would, asking for clarification as necessary.

It does not have a physical embodiment (other than living on a computer screen), but it can control existing physical tools, robots, or laboratory equipment through a computer; in theory it could even design robots or equipment for itself to use.

The resources used to train the model can be repurposed to run millions of instances of it (this matches projected cluster sizes by ~2027), and the model can absorb information and generate actions at roughly 10x-100x human speed⁵. It may however be limited by the response time of the physical world or of software it interacts with.

Each of these million copies can act independently on unrelated tasks, or if needed can all work together in the same way humans would collaborate, perhaps with different subpopulations fine-tuned to be especially good at particular tasks.

We could summarize this as a “country of geniuses in a datacentre”.

Extracted from a talk by Dario Amodei, CEO of Anthropic, an artifical intelligence company, October 2024

This is part of an amazing post by Daniel Amodei, alumnus of Princeton University, Stanford University and the California Institute of Technology and formerly vice president of research at OpenAI before he left to found Anthropic with his sister. Founded in 2021 Anthropic already has 500 employees and is hiring at a rapid rate. The company has already received, or had committed, $6bn of investment from Amazon and Google.

Anthropic expects annual revenue around $500m in 2024 and is valued around $20bn. This is a multiple of 40x sales which is not dissimilar to Palantir’s valuation. Anthropic appears to be growing incredibly rapidly with estimated sales of $3.4m in 2021, $7.9m in 2022, $45m in 2023 and estimates of as much as $850m annualised estimated by the end of 2024.

I am interested in Anthropic, currently unquoted, because it has just entered an important partnership with Palantir and Amazon Web Services. As you can imagine I expect that as and when Anthropic has its IPO there will be considerable excitement, possibly affecting the whole sector.

Near the end of his talk Amodei had this to say.

If all of this really does happen over 5 to 10 years—the defeat of most diseases, the growth in biological and cognitive freedom, the lifting of billions of people out of poverty to share in the new technologies, a renaissance of liberal democracy and human rights—I suspect everyone watching it will be surprised by the effect it has on them. I don’t mean the experience of personally benefiting from all the new technologies, although that will certainly be amazing. I mean the experience of watching a long-held set of ideals materialise in front of us all at once. I think many will be literally moved to tears by it.

Amodei, Anthropic, October 2024

Share Recommendations (28 November 2024)

Palantir PLTR

Amazon (owner of Amazon Web Services) AMZN

Applovin Corporation APP

As recently as October 2008, Amazon was valued at $18bn, similar to the value of Anthropic today. Fast forward 14 years and Amazon is valued at $2.16T. Not only does this present a dramatic picture of how fast values can move in the technology space but I suspect things are moving even faster in the 2020s.

Applovin Exudes Excitement

Last quarter, I shared our confidence in achieving 20pc to 30pc year-over-year growth for the foreseeable future.

We continue to expect 4pc to 5pc quarterly growth through self-learning and market growth, with occasional step changes resulting from enhancements to our AXON algorithm. This quarter, we saw one of those step changes, with meaningful growth driven by advancements to AXON. While we can’t predict the timing of these breakthroughs, we’re in the early stages of AI software development, both within our company and in the broader industry. We expect ongoing research advancements to continue driving our technology forward.

While we remain confident in 20pc to 30pc growth for mobile gaming advertisers alone, we’re also exploring new areas, as shown by our recent e-commerce pilot. Early data has exceeded our expectations, with the advertisers in the pilot seeing substantial returns, often surpassing those from other media channels, and in many cases, experiencing nearly a 100pc incrementality from our traffic. We’re increasingly confident this vertical will scale significantly in 2025 and become a strong contributor for us over the next year and beyond. To support this, we’ve streamlined resources and are reallocating talent from other initiatives to our e-commerce go-to-market team.

In the next few quarters, we’ll launch a self-service platform, opening global opportunities for advertisers of all sizes.

Adam Foroughi, co-founder & CEO, Applovin, Q3 2024, 6 November 2024

Applovin is trading at around 24 times likely 2024 sales so what I would call reassuringly expensive.

Applovin looks like a fantastic company.

We had another fantastic quarter, with total revenue reaching $1.2bn and adjusted EBITDA of $722m , achieving a 60pc adjusted EBITDA margin. This marks a 39pc increase in revenue and a 72pc increase in adjusted EBITDA from the same period last year, translating to an impressive 91pc flow-through from revenue to adjusted EBITDA. In the third quarter, we generated $545m in free cash flow.

That’s up 182pc year over year. Quarter over quarter, our free cash flow grew 22pc, consistent with adjusted EBITDA growth over the same period, and representing 76pc flow-through from adjusted EBITDA to free cash flow. At the end of the third quarter, we had $568m in cash and cash equivalents, and 335m shares outstanding. During the quarter, improvements in our AXON technology, driven by ongoing self-learning and enhancements by our engineering team, contributed to further growth of our Software Platform.

This segment generated $835m in revenue and $653m in adjusted EBITDA, achieving a 78pc margin and growing 66pc in revenue and 79pc in adjusted EBITDA from the same period last year. Quarter-over-quarter flow-through from revenue to adjusted EBITDA was 107pc. Our Software Platform adjusted EBITDA this quarter was positively impacted by several non-recurring cost benefits associated with the renewal of our Google Cloud contract, PSU vesting in the prior quarter, and foreign exchange. On a normalised basis, we estimate our quarter-over-quarter flow-through would be approximately 100pc.

Matt Stumpf, CFO, Applovin, Q3 2024, 6 November 2024

There is an important ‘something new’ hiding at Applovin.

So e-commerce is still in pilot, as we touched on last quarter. I’ll say before jumping into impact this quarter, it’s a super compelling product. Our team has done an amazing job building it.

In all my years, it’s the best product I’ve ever seen released by us, fastest growing, but it’s still in pilot. So compared to the scale of our business inside gaming, it’s too early for e-commerce to make a financial impact that’s material. The step-up this quarter was entirely on the gaming side, which obviously is the vast majority of our advertising business today. We’ve talked for multiple quarters now about how there’s a long runway in gaming.

The gaming category has a need for more UA dollars spent, but it’s constrained by the return on ad spend goals of the advertisers and the technology capabilities that a platform like us has today. Now, our technology continues to improve. We’ve got a long roadmap of enhancements that we can deliver to this platform. Because like I said in the talk script, these AI technologies are just really, really early in existence, both internally and externally.

And all research advancements are going to let us really expand the business inside the gaming category. E-commerce, on the other hand, is looking so strong that it’s something that we think will be impactful to the business financially ’25 and then for the long term.

Adam Foroughi, co-founder & CEO, Applovin, Q3 2024, 6 November 2024

In Q2 2024, Foroughi took some time to explain what Applovin was all about.

Our platform is entirely performance-based. In other words, gaming advertisers who market on our platform generate a measurable revenue and profit from the dollars they spend on our platform. Our customers run marketing campaigns with target return goals but tend to have a much higher appetite for spend on our platform than we can deliver today. And why can’t we deliver more today? Because our current system can only find a limited number of users who will meet their revenue goals.

As our technology improves, we will continue to find more users who achieve these goals, increasing advertiser spend, resulting in materially higher growth than the growth rate of the mobile gaming market. Last quarter, I talked about a goal of growing our software business 20pc to 30pc for the long term. I typically don’t communicate externally about our goals if I don’t have confidence in it. I’m communicating it now because I do have strong confidence in it, and I see many years of growth ahead of us.

Here are the primary drivers of that growth goal. Continued improvement from our models as they learn from more data. As our models gather more data, they’ll become more accurate and find more good users for our advertisers, gains that our team delivers to the efficacy of our models through enhancements. Our research, science, and core engineering team members are exceptionally talented and consistently deliver lifts to the performance of our models.

Demand expansion into new verticals, we just launched the first web advertising campaigns for shops this quarter. And while in pilot right now, we think it will unlock a lot of demand expansion opportunities for us. And then supply expansion, as we broaden out our demand base outside of gaming, we expect the new categories will really help grow our CTV [connected TV] footprint. We continue to be very excited about our prospects and the performance our team is able to deliver.

Adam Foroughi, co-founder & CEO, Applovin, Q2 2024, 7 August 2024

Strategy – When You Find Exciting Shares Buy Some

I think subscribers won’t be surprised to learn that the two which excite me most here are Palantir and Applovin. Amazon.com is more mature as an investment and less focused on what is most exciting in the technology world of 2024.

A bold investor could consider both Palantir and Applovin for $-cost averaging programmes, maybe even with a little leverage.