ExoBrain
AI adoptioncompute infrastructureenterprise AIfrontier labs

A tale of two cities

The piece analyses divergent market reactions to AI investment strategies, contrasting Meta's heavy capital expenditure with the successful partnerships of Microsoft and Google, while noting the rising costs of frontier research.

Joost de Jonge

Joost de Jonge

3 min read
A tale of two cities

This week, the AI industry experienced a mix of signals as major players saw varying fortunes from their investments and strategies. BCG, Microsoft, and Google demonstrated positive results, while Meta faced investor concerns over its AI expenditure.

BCG’s announced it expects 20% of its 2024 revenue to be AI-related. 2023 revenue was $12.3 billion. This signalled a positive shift in the world of strategy consulting. The firm is reimagining how it delivers its services, leveraging AI to transform the way it advises clients. But is this a sustainable strategy, or will strategy consultants eventually need to invest more heavily in their own AI capabilities to stay competitive?

As the Stanford report revealed last week, staying at the frontier of AI research is becoming increasingly expensive. Training costs for cutting-edge models like GPT-4 and Gemini Ultra are estimated to be in the hundreds of millions of dollars. This has effectively priced out organisations such as universities, once the centres of ground breaking AI research, from developing their own frontier foundation models. Policy initiatives, such as President Biden’s Executive Order on AI, aim to level the playing field by providing non-industry actors with the resources needed to conduct high-level AI research. However, it is unlikely these efforts are enough to counterbalance the immense resources of tech giants.

Meta’s $40 billion capital expenditure on AI, with expectations of spending $100 billion in 2024, spooked investors and wiped out $120 billion in company value in a single day this week. While the scale of Meta’s investment is undoubtedly massive, it’s worth noting that the company has poured $46 billion into the metaverse over the last three years, and Amazon invested over $20 billion in Alexa. These bets, however, seem more binary in nature compared to the broader potential of AI.

Microsoft, on the other hand, has taken a different approach by partnering with OpenAI. This strategy seems to have gone down better with investors, as evidenced by Microsoft’s position as the world’s most valuable listed company, with a valuation of nearly $3 trillion and a climbing share price following stellar last quarter cloud earnings. The symbiotic relationship between Microsoft and OpenAI showcases the potential for strategic partnerships in the AI space, allowing companies to share the risks and rewards of cutting-edge research. Google also signalled last week that it would be spending $100 billion on AI in the coming years, although direct AI revenues from cloud services are sending Alphabet’s share price in the upward direction.

Takeaways: The AI industry is at a critical juncture, with fortunes being made and lost in the blink of an eye. Companies must carefully consider their investment strategies and partnerships to navigate this rapidly evolving landscape. Investor sentiment remains jumpy, but the opportunity presented by AI is too big to ignore and talk of a ‘bubble bursting’ is premature. As the dust settles on another busy week, Amara’s law is as relevant as ever; we tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run. The cost of AI feels out of step with the immediate benefits for some, but the longer-running growth signals suggest AI’s vast potential.