In recent years, investment in artificial intelligence has accelerated dramatically, reaching hundreds of billions of dollars. Will it pay off?
Big companies succeed. Small companies struggle. That certainly affects the return on investment.
Over the past 10 years, investment in artificial intelligence has accelerated by leaps and bounds, reaching hundreds of billions of dollars.
- Much of the investment in AI infrastructure is future-oriented.
- Outside of OpenAI, Claude and a few others, consumer adoption of AI technology is limited.
- Businesses are also at an early stage in the adoption curve for this technology. Large-scale adoption is still on the horizon.
However, the economic benefits have not yet matched the investments. Emerging technological advances, such as large language models, have yet to be fully adopted in most companies, and while the technology has seen one of the fastest adoption curves, it is currently expensive to develop.
This pattern is typical of emerging technologies. For example, sequencing the human genome initially cost $1 billion, whereas it now costs about $100 million.
While OpenAI has surpassed $3 billion in revenue, many other AI companies are struggling to break the $100 million barrier. The current market is largely focused on developing foundational frontier models and technologies, which enable products such as AI companions like Friend.
In the AI “wrapper” space, where startups build products around AI lab APIs, competition is fierce. These companies often struggle to break the $100 million revenue barrier, even as they hone their models for specific use cases.
A major risk is the emergence of new AI models that can inherently perform these specialized tasks, potentially rendering these startups’ refined solutions obsolete.
For example, when ChatGPT launched, jasper.ai lost subscribers, leading to staff cuts, and copy.ai now operates in an extremely saturated market. This challenge underscores the volatility and rapid evolution of the AI industry, which makes it difficult for smaller companies to achieve significant market traction and differentiation.
This big-vs-small situation creates a significant gap between major players like OpenAI, MidJjourney, and Anthropic and the rest of the industry. This is because consumer adoption of AI technologies is limited, outside of a few key products like Claude, ChatGPT, MidJourney, and Runway.
However, these models are capital intensive to run, with rumours suggesting that ChatGPT costs a staggering $700,000 a day to operate. That’s not counting all the staff and R&D and training costs for the new models. The high costs and investments exclude many companies from competing.
This has raised concerns in the market and led to some rumors, one of which is that OpenAI could run out of cash within a year. While this seems unlikely, the company needs to continue attracting investment and expanding its operations to have a clear path to profitability. But that is not the goal for now.
Because AI development is still in its early stages, companies like Microsoft, Amazon, and Google are leading the charge with substantial investments in AI and data center infrastructure.
The venture capital ecosystem is also very active in AI investments. Firms such as Sequoia Capital and Andreessen Horowitz are among the most active and prominent investors in the AI space, especially in generative AI startups.
Investments in infrastructure ensure that AI labs can stay ahead of the curve, introducing the latest models and staying competitive. Building this infrastructure is crucial for the future, as it enables the development and deployment of even more advanced AI technology.
One of the biggest components of infrastructure is computing, with investments potentially reaching a staggering $1 trillion over the next few years. Major tech companies such as Microsoft, Google, and Amazon are investing heavily in this sector, with each data center costing around $2 billion to build. The field is still nascent, as companies are just learning how to set up these GPU-specialized data centers.
These centres will be equipped with the latest chips, such as the H100. However, these chips will quickly become obsolete as more powerful ones emerge, requiring continuous reinvestment to meet the increasing computational demands of new AI models.
While it can be argued that certain labs have advantages in models, algorithms or data, competing in this space is challenging. Researchers often move between AI labs, transferring knowledge and reducing competitive advantages.
One of many examples is Dario Amodel, former VP of Research at OpenAI, who co-founded Anthropic in 2021. When it comes to making money on invested capital, what are AI labs and their investors really betting on?
Although AI is not yet on every company’s roadmap, AI labs are counting on reducing the cost of intelligence and its value to companies that want to acquire it.
Companies are currently investing heavily in recruiting top talent, which is a significant expense. While current AI models resemble clumsy interns or junior employees, they are improving and becoming cheaper.
For example, OpenAI’s GPT-4o-mini is 97 percent cheaper for input tokens and 96 percent cheaper for output tokens, compared to GPT-4. This reduction translates to a 97 percent reduction in cost. This reduction translates to a 97 percent decrease in the cost of a clumsy intern’s intelligence. Imagine if this intelligence reached PhD-level capabilities — the implications for cost savings and efficiency would be immense.
In the near future, digital workers, also known as AI agents, will collaborate with humans and other AI agents. At first, they will automate mundane tasks, but over time they will take on higher-value activities.
This shift could allow humans to focus on more important problems, potentially reducing the need for so many human workers. Smaller groups of humans, supported by thousands of digital agents handling non-strategic tasks, could produce more valuable results and tackle complex problems more effectively.
When considering investment in AI, one might ask whether capturing only a small portion of the tasks that humans perform today will yield significant returns on investment. This pattern is typical of emerging technologies. For example, sequencing the human genome initially cost $1 billion, whereas it now costs about $100.
Similarly, AI has the potential to transform several industries by lowering the cost of intelligence, thereby creating significant economic value and improving overall productivity.
