AI's God Delusion
AI's God Delusion
Why our digitally enhanced future will be emergent & how investors need to stay in the game to benefit from its uncertainty
After covering Mike Lynch's improbable life and death, I continue with the themes of nautical uncertainty and AI.
I recently chatted with London-based Israeli software entrepreneur and generative AI (GAI) evangelist Ami Daniel, co-founder and CEO of AIM-listed Windward. His hands-on view of GAI's benefits got me thinking. So, with the much anticipated Nvidia earnings release, it seemed a good time to riff a bit on how investors should think about our digitally enhanced future and its potential risks.
Aged 18, Ami joined the Navy and became a deck officer responsible for landing helicopters on Corvette class (small, fast) ships. In 2006, while serving on INS Hanit, his ship was hit by a Chinese missile fired by Hezbollah from Lebanon. It struck the vessel on the water line. The flight deck and propulsion system were knocked out, four of its sixty-four crew members were killed, with a further dozen injured. Following this life-defining near-death experience, Ami vowed to start a company with a mission to solve what he calls the small ship, big ocean problem.
Fourteen years later, Windward is valued at £120m in the London AIM market. While still an immature business, Ami believes it can one day become a multiple £100m’s revenue, Rule of Forty company. Governments and businesses can track, manage, and protect their maritime assets worldwide using Windward's expanding range of products, which increasingly feature AI enhancements.
Ami has energy and passion and is adamant that GAI will permanently and positively change our lives in a bottom-up, problem-specific manner. What is more, he can provide real-world revenue-generating examples.
He said that
I know some people are a bit cynical about Generative AI. In my view, they're cynical from a macro perspective. They're not cynical from a practical product perspective, which is very different. They're cynical in terms of, "Oh my God, how will Microsoft and the Magnificent Seven make $400 billion more to make up for their valuations?" I don't know, I don't care. I don't need to make $400 billion. I just need to triple my company. So the question is, will it help you triple your company faster? And the answer is absolutely yes.
For him
When the world changes, you either change with it or you die; Nokia died. I almost died once. I have no intention of doing it again. Not from a business perspective or personal perspective...Generative AI is a revolution. We're not there yet, but if you think a few years into the future, it will be a revolution and benefit the people with proprietary data.
Windward has a new GAI product called MyExpert. Ami says
People love it. They're already buying it. I was surprised how fast they're buying it. It opens up a whole lot of really, really interesting automation conversations with our customers. It just sparks their imagination, and they say, "Oh my God, you can do this? What else can you automate for my business?" So that's cool. But the bigger companies will benefit more because they have more data.
For Ami, his customers have data that they need to use more effectively in their businesses. They need someone to structure their unstructured data, and he wants that person to be him. I didn't want to remind him that the last tech entrepreneur who made this claim to London investors came to a tragic, watery end only two weeks ago.
In a podcast released last week, former Fidelity tech investor Gavin Baker gave one of the best rundowns I've ever heard on the outlook for AI and its potential impact on the investing landscape. In the week when Nvidia's top-of-the-range results couldn't match investors' giddy expectations, Baker gave a brilliant assessment of what is driving the investment boom in Nvidia products.
Mark Zuckerberg, Satya, and Sundar just told you in different ways that they are not even thinking about ROI. And the reason they said that is because the people who actually control these companies ... believe they're in a race to create a Digital God.
And if you create that first Digital God, we could debate whether it's tens of trillions or hundreds of trillions of value. We can debate whether or not that's ridiculous, but that is what they believe, and they believe that if they lose that race, losing the race is an existential threat to their company.
Larry Page has evidently said internally at Google many times, "I am willing to go bankrupt rather than lose this race." So everybody [else] is really focused on this ROI equation, but the people making the decisions are not because they so strongly believe that [AI’s] scaling laws will continue.
So there you have it: the supernormal investment cycle in Nvidia chips and hyperscale data centres is driven by game theory, not normal ROI decisions. As Bernstein says, game theory tells us that the true source of uncertainty lies in the intentions of others.
However, Baker, quoting Tiger Global's Chase Coleman, points out that we are still very early in the journey to an AI-enhanced future.
He [Coleman] had a really interesting statistic: ChatGPT came out in 2022 and was to AI as Netscape Navigator was to the internet in 1994. Less than 1% of the current global internet market cap was founded in the two years after Netscape Navigator came out.
Baker sees more hardware layers (such as storage and networking products) as ripe for next-stage AI investments. He considers Windward's and the thousands of other application layers much more complicated prospects and likely to emerge much later. He may well be right. What is clear is that big players are driving this market with huge balance sheets that are not rational actors, a backdrop for greater market volatility and uncertainty. As if investors don't have enough to worry about already?
Governments and their central banks are the only entities with balance sheet capacity in the same league as the big tech stocks, and they are currently changing their focus on the critical issues of their concern. As Fed Chair Powell said in Jackson Hole, their focus has turned from inflation to the jobs market. Yet, no sooner had he signalled this critical strategic move than the US Bureau for Economic Analysis (BEA) revised its estimate for Q2 GDP growth from 2.8% to 3.0%. Surely, inflationary?
While relying on short-term economic measures is not advisable, the US economy is growing faster, employing fewer people, and has the highest interest rates since 2007. What other than faulty statistics could be causing this? Of course, one answer is the adoption of AI. Too early for this, perhaps? Well, the buy now pay later pioneer Klarna suggests otherwise.
In February, the company reported that its AI chatbot based on OpenAI IP significantly benefited its business.
The AI assistant has had 2.3 million conversations, two-thirds of Klarna's customer service chats
It is doing the equivalent work of 700 full-time agents
It is on par with human agents regarding customer satisfaction score
It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries
Customers now resolve their errands in less than 2 mins compared to 11 mins previously
It's available in 23 markets, 24/7 and communicates in more than 35 languages
It's estimated to drive a $40 million profit improvement to Klarna in 2024.
In May, it reported that:
9 out of 10 Klarna employees now harness the power of Generative AI in their daily work.
This transformative integration spans all teams, with non-technical groups such as Communications, Marketing, and Legal seeing adoption rates of 93%, 88%, and 86%, respectively. These figures demonstrate AI's versatility and role in enhancing non-technical operations, a strategic move that sets Klarna apart.
This internal AI revolution is further propelled by Kiki, Klarna's bespoke internal AI assistant, which has adeptly responded to over 250,000 inquiries (2,000 per day) since launching in June 2023, establishing itself as an indispensable asset within the company. Overall, 85% of all Klarna employees are now using Kiki.
In recent days, Klarna CEO Sebastien Siemiatkowski has said that its workforce has shrunk from 5,000 last year to 3,800, with a further reduction to around 2,000 expected in the coming years.
We should put these statements into context: this loss-making business with significant consumer credit exposure is preparing for an IPO. However, even with a conservative interpretation of what GAI can do to Klarna's operating metrics, multiplied across all customer service cost bases, this represents a material labour productivity improvement and a significant deflationary macroeconomic risk factor.
So, what should investors be worried about most: the return of inflation, debt default, recession, and mass unemployment? Trying to understand the world from a macroeconomic perspective can be like playing 4D chess.
Fortunately, last week, two investment gurus, Howard Marks and Morgan Housel, discussed how investors should think about this radically uncertain outlook.
Housel repeated his great line,
The more debt you have, the narrower the range of volatile outcomes you can endure in life. You get paid for two things when investing. One is mispricing; the other is the endurance of volatility. When you view debt as a narrowing of the outcomes you can endure, it takes on a new level of significance.
Virtually every investor says, "I'm a long-term investor." At least they aspire to be a long-term investor. But that means is you have to be willing to forego shorter-term gains. And that is much more difficult to do. It's easy to say, "I'm in it for the long term." It's much harder to say, "I'm going to go the next one or five years below my potential."
Then Housel recounted the story of the third founder of Berkshire Hathaway, Rick Guerin. Buffett said Rick was just as smart as us, but he was in a hurry.
Rick had margined much of his Berkshire stock and enjoyed some spectacular years during the 1960s. Then 1974 happened, and Rick was sunk. Warren bought him out at pennies on the dollar, never to be heard of again.
As expectations rose ahead of the Nvidia earnings report last week, investors gathered in eager anticipation in Manhattan bars at formally arranged earnings call parties. And the bulls were not to be disappointed with the chip giant's business outcome.
Yet, as Peter Bernstein said, vast ills follow a belief in certainty. Or Mark Twain more commonly related, It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so.
The Nvidia share price fell and remains 15% below its June high. Its foreseeable future remains attractive, but the market for its equity has, for now, run out of people to buy it. The big tech titans, driven by their perceived existential threat, must own tens of thousands more Nvidia devices to achieve their dreams of winner-take-all superintelligence. However, the scaling laws Gavin Baker discusses might stop. Their plans will eventually collide with economics.
As Peter Bernstein told us in 1996,
The real trouble with this world of ours is not that it is an unreasonable world, nor even that it is a reasonable one. The commonest kind of trouble is that it is nearly reasonable but not quite. Life is not an illogicality, yet it is a trap for logicians. It looks just a little more mathematical and regular than it is; its exactitude is obvious, but its inexactitude is hidden; its wildness lies in wait. But therein lies the logician's trap: past data from real life constitute a sequence of events rather than a set of independent observations, which is what the laws of probability demand. Life is a collection of similarities rather than identities. No single observation is a perfect example of generality.
Investors should remember that stock markets are short-term voting machines and long-term weighing machines. The assessments we are voting on already have somebody else’s expectations priced in, and we never know the exact weight but rather witness a battle for its determination.
Bernstein again,
In December 1972, Polaroid was selling for 96 times its 1972 earnings, McDonald's was selling for 80 times, and IFF [International Flavors & Fragrances] was selling for 73 times; the Standard & Poor's Index of 500 stocks was selling at an average of 19 times. The dividend yields on the Nifty-Fifty averaged less than half the average yield on the 500 stocks in the S&P Index. The proof of this particular pudding was surely in the eating and a bitter mouthful it was. The dazzling prospect of earnings rising up to the sky turned out to be worth a lot less than an infinite amount. By 1976, the price of IFF had fallen 40%, but the price of US Steel had more than doubled. Figuring dividends plus price change, the S&P 500 had surpassed its previous peak by the end of 1976, but the Nifty-Fifty did not surpass its 1972 bull-market peak until July 1980. Even worse, an equally weighted portfolio of the Nifty-Fifty lagged the performance of the S&P 500 from 1976 to 1990.
The former Economist editor and man of letters, Walter Bagehot, said, "People are most credulous when they are happy." Nvidia investors have had much to be happy about and will have many better days to celebrate their amazing company.
However, the less risky beneficiaries of GAI will emerge from below in verticals ploughed by specialists like Ami at Windward, maybe Klarna, and many more people searching for AI's secret benefits. Their path to success and impact on the broader economy are uncertain but will significantly impact in aggregate and over time. You require style and asset diversification to optimise your chances in this scenario. Remember, though, diversification doesn't stop you from losing money; it just stops you from losing it all at once.
Bon chance.
Jeremy