It is estimated by Sequoia Capital that by the end of 2024 the investment in AI Infrastructure (Processors, Data centers and Software) will have reached $600 Billion in just 2 years. That is equal to the GDP of Sweden and Argentina.
Why it matters
The AI Chatbot “ChatGPT” has become one of the fastest growing software platforms and companies (OpenAI) in tech history. Today, it has over 200 million monthly users that use the AI technology to write content, research and create AI assistants that enhance our minds and intelligence. Maybe we should give artificial intelligence a new moniker –“EI” – Enhanced Intelligence.
Because of the remarkable success and rapid growth of the simple AI interface (Not unlike the Google search interface) launched by ChatGPT in November 2022, the global AI industry has exploded, with many using it as the chatbot answer engine.
Global companies like Google, Facebook, Amazon, Nvidia and Microsoft have created new products and invested in the AI infrastructure.
We have also seen new startups launched and backed by venture capital firms. Investors have also seen this as a new gold rush and have opened up their wallets and cheque books. Greed doesn’t mind a new hook to hold onto.
If this is a bubble and it collapses we could see financial wreckage at a scale like the 2008 GFC and the 2000 dot-com crash. This could destroy the investment of tech firms, VC companies, take down AI startups and destroy the investment portfolios of the individuals backing this trend.
By the numbers
The scale of the money pouring into AI is unprecedented. The attention is building. The big boys (The magnificent seven) and the Venture Capital Funds are piling in with the big dollars. And the world is having many conversations online about the future of artificial intelligence.
So what are some of the numbers to provide some perspective on its scale and speed of adoption?
The FANG cohort that has become the “Magnificent 7” is first cab off the rank. They have the capital, profitability and balance sheet to invest. Then there are the Venture capital companies who are trying to pick winners.
FANG (Facebook, Amazon, Netflix and Google)
FANG was a term that emerged that valued these global companies as trillionaires (as a stock market valuation measurement). But now we need to add the “Magnificent 7” to a bigger group. This muscles up the group from four to seven. Adding Microsoft, Tesla, Nvidia and Apple and Tesla is appropriate as they are all now Trillion dollar companies.
Selling dreams is an Elon superpower. But is Tesla is maybe an overrated dream that could become a nightmare. It has now fallen off the trillion dollar list.
The investment numbers attributed to AI from this exclusive club is approximately $600 billion that has been invested into AI infrastructure in the last 2 years.
Venture capital
Rising interest rates hobbled funding and those cheap funds dried up for most startups.. Despite this, the resurrection of investment into startups is now driven by AI.
In the last quarter, nearly half the US capital investment ($27 Billion) went to AI companies . It is a boom and it is going into the tools and shovels for the AI gold rush.
Working out where the trillion dollar companies money trail of investment is going, if it is worthwhile and will it produce a return needs to be considered in the backdrop of history.
Flashback: Historical bubbles
History reveals that economic bubbles are not new. The Dutch had Tulip Mania and then there was the South Sea Bubble. In the modern era we’ve had the following:
Dot-com bubble (1999-2000)
The Dot-com Bubble, spanning roughly from 1999 to 2000, was a period of extreme growth and subsequent crash in the stock values of technology and internet-based companies.
Here’s a quick summary of its causes and outcomes: Let’s look at the causes and outcomes of this bubble.
Causes
- Rapid Internet Adoption: The rapid increase in internet usage during the late 1990s drove excessive speculation about the potential of tech companies.
- Venture Capital Funding: There was a significant influx of venture capital into startups, many of which had unproven business models and little to no revenue.
- IPO Frenzy: Many companies went public with initial public offerings (IPOs) that saw their stock prices soar on the first day, often without solid fundamentals to justify the high valuations.
- Speculative Investments: Investors, driven by fear of missing out (FOMO), poured money into tech stocks, inflating their values beyond reasonable economic fundamentals.
- Media Hype: Extensive media coverage contributed to the frenzy, hyping the potential of the internet and new tech startups.
Outcomes
- Market Crash: The bubble burst in 2000, leading to a sharp decline in stock prices. The NASDAQ Composite, heavily laden with tech stocks, fell dramatically from its peak.
- Bankruptcies and Closures: Many dot-com companies went bankrupt or ceased operations, leading to significant job losses.
- Economic Slowdown: The burst of the bubble contributed to a broader economic slowdown and was a factor in the early 2000s recession.
- Regulatory Changes: The bubble and crash led to calls for and implementation of tighter regulatory controls over IPO practices and venture capital funding.
- Shift in Business Models: Post-bubble, surviving tech companies shifted towards more sustainable business models, emphasizing revenue, profit, and real business metrics over mere user acquisition and speculative growth.
The Dot-com Bubble serves as a cautionary tale about the dangers of speculative excess, particularly in emerging technological sectors.
Housing bubble (2006-2008)
The Housing Bubble, which peaked around 2006 and burst in 2008, was a period of rapid escalation in home prices followed by a severe collapse, leading to the global financial crisis.
Here’s a brief summary of its causes, outcomes, and some key statistics:
Causes
- Low Interest Rates: Post-9/11, the Federal Reserve lowered interest rates to stimulate the economy, making mortgages cheap and encouraging borrowing.
- Subprime Lending: Banks issued more subprime mortgages to borrowers with poor credit histories. By 2005, about 20% of all U.S. mortgages were subprime.
- Securitization of Mortgages: Financial institutions bundled mortgages into securities sold globally, spreading and obscuring the risk.
- Speculative Buying: Many purchased homes as investments, assuming prices would continue to rise, which inflated the bubble further.
- Lax Regulation: Inadequate oversight allowed risky lending and borrowing practices to proliferate.
Outcomes
- Housing Market Crash: Home prices plummeted by about 30% from their 2006 peak, marking the most significant price drop since the Great Depression.
- Foreclosures: The U.S. saw a dramatic increase in foreclosures, with nearly 3 million foreclosures filings in 2009 alone.
- Financial Crisis: The collapse triggered a global financial crisis. Major financial institutions faced bankruptcy, leading to extensive government bailouts.
- Recession: The crisis led to the Great Recession, with U.S. GDP contracting and unemployment reaching 10% in October 2009.
- Regulatory Reforms: It resulted in major regulatory changes, including the Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010.
Key statistics
- Peak to Trough Home Price Decline: National home prices dropped approximately 30% on average from their peak in 2006 to the trough in 2011.
- Foreclosure Rates: At the height of the crisis in 2010, one in every 45 homes in the U.S. received a foreclosure notice.
- Financial Sector Impact: The crisis wiped out nearly $11 trillion in household wealth in the U.S. between 2007 and 2009, primarily due to declines in stock and home values.
The Housing Bubble’s burst had profound and long-lasting impacts on the global economy, demonstrating the interconnectedness of modern financial systems and the dangers of excessive risk-taking in the financial and housing markets.
Is the AI Investment Boom and the Dot-com bubble the same disaster waiting to happen?
The AI investment boom shares some similarities with the Dot-com bubble, such as the rapid increase in capital investment and high expectations for new technologies, but there are several key differences that set them apart:
Maturity of the Companies: Unlike the Dot-com era, many of the companies leading today’s AI boom are established and profitable entities. Companies like Nvidia, Microsoft, and other tech giants have robust financial foundations, which contrasts with the many unprofitable startups that fueled the Dot-com bubble (CCN).
Valuation and market metrics: The valuation metrics during the Dot-com era reached extremely high levels, with the Nasdaq 100’s forward price-to-earnings ratio peaking at around 60.1 times in March 2000. Today, despite high valuations in tech, these ratios are more conservative, reflecting a market that places greater emphasis on earnings and sustainability (Visual Capitalist).
Technological and market readiness: AI technology is being integrated into a variety of industries and is supported by substantial improvements in infrastructure, such as cloud computing, which were not as developed during the Dot-com era. This integration is supported by significant investments from large tech companies, ensuring a broader and more immediate applicability and impact (CCN).
Investment nature: The current AI investment landscape relies less on debt financing and more on venture capital and reinvestment of substantial cash flows from profitable operations. This differs from the Dot-com era, where much investment was fueled by speculative equity investments and debt, leading to unsustainable financial structures (CCN).
Regulatory and market environment: Post-Dot-com, there’s been a tightening of regulations and a more cautious approach from investors towards new technologies. This has led to a more measured expansion in tech investments compared to the unchecked speculation of the late 1990s (Visual Capitalist).
These differences suggest that while there are concerns about an AI bubble similar to the Dot-com bubble, the underlying financial health of key players, the advanced state of technology integration, and more cautious investment practices provide a buffer against a similar bust.
So, AI as a bubble?
Mustafa Suleyman, who was one of the co-founders of DeepMind that was founded in the UK in 2010 and was acquired by Google in 2014 for $400 million+, is the poster boy of AI. In October 2015, a computer Go program called AlphaGo, developed by DeepMind, beat the European Go champion, Fan Hui. This was the first time an artificial intelligence (AI) had defeated a professional Go player.
But AI investment success is not easy even for the successful few and avoiding the bubble is even harder. In 2019, he left DeepMind and joined Google.
He left Google in 2022 and founded “Inflection” and raised $1.3 billion to build a chatbot in a similar vein to ChatGPT, called Pi, which ran on its own AI model to provide a personal AI. It claimed Pi could be a coach, confidante, creative partner, sounding board and assistant. Despite his experience and expertise, he wasn’t able to avoid failure.
The signs pointing to a bubble of overconfidence can lead to hubris and speculative investments. Despite raising the $1.3 billion, Inflection never realized significant revenue and was absorbed into Microsoft.
Quick returns are not what is at play here.
Despite investors wanting to find rich gold seams tomorrow.
AI’s deep integration across various sectors and long-term potential is a long game. And patience is required.
Patience
To create some perspective on the long game you only need to look at some of the other technology companies’ profit journeys since the 1980’s. Let’s take a look at Facebook and Amazon.
Facebook, founded in 2004, reached profitability in 2009. The company’s path to profitability was primarily fueled by its rapidly growing user base and its ability to attract advertisers eager to reach its audience. Facebook’s strategy focused on expanding its platform to a broad audience by initially connecting college students before opening up to the general public. By 2009, Facebook had grown substantially in terms of users, which made it an attractive platform for advertisers. This surge in advertising revenue was crucial in helping Facebook achieve profitability within five years of its founding.
Amazon
Amazon, founded in 1994 by Jeff Bezos, took quite a bit longer to become profitable compared to many other tech companies. It reported its first full-year profit in 2003. This was nearly a decade after its founding. Amazon’s initial focus was on rapid growth and expansion into various markets, such as books, electronics, and other consumer goods, often at the expense of immediate profitability. The company reinvested most of its revenue back into expansion and infrastructure, such as distribution centers and technology, which delayed its profitability but set the stage for its dominant position in the online retail market.
So wanting a return in a couple of years is a bit immature and impatient,
In summary Facebook took 5 years to get a small return on its investment and Amazon took a decade!
What are the experts saying?
The experts who have vast industry experience and expertise often struggle with the future. IBM’s Tom Watson famously said. “I think there is a world market for maybe five computers.” Predictions are everywhere. And the truth is nowhere.
No one predicted the 2000 dotcom and housing crash and the GFC architects were hidden from view. Chasing the latest trend or fad can lead to tears.
Active investment advisory agencies have been revealed to produce a poorer return than just passively investing in the indexed funds. Set and forget beats fiddling.
So in essence, we need to listen to the advice of Charlie Munger (Of Berkshire Hathaway fame and the right hand man of Warren Buffet), who passed away last year at the age of 99, just 3 months short of 100 years.
He had this to say about investing:
“The big money is not in the buying and selling … but in the waiting”.
Get rich quick schemes are enticing. But they are very tempting but often a scam.
What’s next?
Artificial intelligence has arrived. But it is evolving fast and we as humans evolve slowly. Charles Darwin premised the evolution journey in the millions of years. Expecting change tomorrow is a fool’s errand.
So…Lean into the future but don’t demand. Do not force fast solutions. Or expect a quick return. AI will change your world. One prompt at a time.
You have access to the “World Brain”. But that doesn’t mean you have wisdom. That takes time, experience and pain. We don’t learn from comfort.
Embrace discomfort. It’s life’s teacher. So watch the world. Read and reflect. And prompt the AI “Oracle” to look for wise answers. Absorb the world’s intelligence. Experiment with the AI trend.