In 1507, German cartographer Martin Waldseemuller’s Universalis Cosmographia was published. It is the first known European map to incorporate the “discovery” of the Americas. Despite 15 years having passed since Columbus sailed the ocean blue, little was actually known of the New World at that point, and Waldseemuller labeled the land to the west of the Atlantic as Terra Incognita, or Unknown Lands.
The discovery of the New World caused a mix of excitement for potential economic gains and fears over disruption of the old order. Terra incognita, indeed.
AI has been part of the computer science landscape for at least seven decades, since the great and tragic British polymath Alan Turing postulated his eponymous Turing Test, which would set the standard for human-machine interaction.
ChatGPT, part of the broader family of artificial intelligence, has dominated both news cycles and financial markets. A quick review of Google trends shows that search popularity for “ChatGPT” and “generative AI” have gone asymptotic, and for good reason—it is the first time consumers can “touch” AI that passes the Turing test. And it’s not just consumers that have been captivated by ChatGPT. Year-to-date S&P 500 gains are largely attributed to a small handful of technology companies with strong AI narratives.
In some ways, ChatGPT and generative AI mirrors the discovery of the New World. Against the excitement of a powerful example of artificial intelligence, there are a vast number of questions, ranging from the promise of potential productivity gains to the potential mass dis-employment. AI is the technologia incognita, the unknown technological frontier of our digital age.
Our many conversations with clients, industry participants, and companies center on three broad unknowns: the impact on the labor market, the impact on companies and industries, and the impact on financial markets.
Every generation has seen new, powerful, and defining technologies emerge, from the new weaving machines that Ned Ludd smashed in protest in 1779 to Thomas Watson Jr.’s IBM System/360 that disquieted office workers in 1964. The centuries change and the technology evolves, but the fears remain constant: machines and automation were going to throw workers out of jobs and bring on a dystopian state of perpetual under-employment.
Nothing could be further from the truth. There were certainly dislocations in industries that had not adapted to new technologies. However, the historical record shows that technology ultimately makes productive workers more productive. The productive contribution of technology is perhaps more important than ever in an economy marked by challenging demographics and high wage inflation.
We have no doubt that some industries and some parts of the labor market will face significant challenges from AI, but at the same time we see this more as a repeat of previous cycles than a paradigm shift.
Looking ahead, we see artificial intelligence as a broad horizontal technology that will impact every industry across every region of the global economy, driven by gains in computer power, systems development, but especially ever-growing amounts of data.
In an AI-driven world, data isn’t the “new oil”, it’s actually much more valuable. Oil is a commodity, fungible with adjustments across regions and grades. On its face, data may appear as a commodity. But we think companies will put AI to work on the most valuable data in the world—their own proprietary data about their customers, their operations, and their finances. Machine learning, neural networks, and large language models will drive returns on data that have been so far left fallow. Who will win in this AI and data-driven arms race? Time will tell, but we think companies that have invested in the infrastructure and operations to capture vast amounts of customer data are in the pole-position.
Lastly, financial markets have clearly championed AI, with the YTD winners having strong AI narratives. Are these gains justified? In the near-term, perhaps not as full realization of the promise of AI is likely an intermediate-term horizon at best. Longer-term, we’re bullish on AI as a transformative technology, but we’re cognizant of the history of technology—most of the darlings of the dotcom era either faded into irrelevance or have lingered in a sort of limbo of low growth and cheap valuation. Will the market leaders of today follow this path? It’s too early to say, just as it is too early to pick the definitive winners in the fast-moving world of AI.
Walderseemuller’s terra incognita eventually became home to over 1bn people and roughly one-third of global GDP. Rather than fear AI and disruption, investors and individual need to embrace the currents of change inexorably flow towards technologia incognita. At the same time, investors need to remember not just the history of technology, but the history of markets. The leaders of one cycle don’t necessarily become the leaders of the next cycle. Some companies are left behind, while new companies emerge. Valuations do not influence returns much in the short-term, but are a key factor in the long-run.
Main contributors: Solita Marcelli and Kevin Dennean
Read the full report Technologia incognita, 23 May 2023.
This content is a product of the UBS Chief Investment Office.