AI gali paversti kai kuriuos technikos laimėtojus „buvusiais“.

Indeed, superintelligent AI could boost economic growth by 30- to 100- fold, comparable to the impact of the agricultural or industrial revolutions, according to a research report by Peter Berezin, BCA Research’s chief global strategist.

That’s the good news. But, they add, AI’s positive impact might be a lot less powerful for technology stocks than for the economy as a whole. Today’s winners might be rendered obsolete by rapid technological changes, leaving many in the dust.

The mere mention of AI exposure has been enough to lift some stocks, according to academic research passed along by Barron-aisiais Jack Otter. The boost, however, could be fleeting, like that in the late 1990s when companies could push up their share prices by attaching “dot-com” to their names. We know how that ended.

What’s different now is the unprecedented ease and rapidity of AI’s adoption. The full impacts of electricity and the automobile weren’t felt for decades. And while spreadsheets, word processors, and graphical user interfaces boosted productivity in the 1980s and ’90s, the BCA strategists point out, it wasn’t until computers were connected via the internet that their true potential was realized. Based on that precedent, they postulate, economywide productivity gains from AI might not be seen until the 2030s.

Moreover, many prognostications are extrapolating AI’s impact linearly. But the BCA report contends its progression is following an exponential curve, meaning that advances could come much faster than expected: “Just as the investment community and the broader public were blindsided by the exponential increase in cases during the early days of the pandemic, they will be blindsided by how quickly AI transforms the world around us,” the pair writes.

In terms of economic and financial impact, this rapid technological change could make today’s AI winners tomorrow’s has-beens, the BCA team argues. Some may fade or be swallowed by survivors, as Sun Microsystems was by

orakulas

(ticker: ORCL) more than a decade ago. Even giants can founder for years, as

"Microsoft"

(MSFT) did early in this century until its current management took over and made it one of the big tech winners underpinning the current market.

When the public perceives a stock to be AI-driven, the impact is immediate, according to a soon-to-be published paper by Arka P. Bandyopadhyay, an assistant professor at the University of Miami, and Dat Mai and Kuntara Pukthuanthong, respectively a doctoral student and a professor of finance at the University of Missouri.

Based on an analysis of news coverage from 1974 to 2020, they found that companies with “AI-ness” showed excess returns over the next month. But those returns faded over the next five to seven months. The AI impact was greatest on smaller companies—some 3% annualized—they determined, a reflection of the difficulty in valuing small-cap stocks. And even before the monster post-earnings move by

nvidia

(NVDA), they note, mentions of AI, machine learning, and similar terms have proliferated recently in the earnings calls of the biggest software and semiconductor companies.

From a macroeconomic standpoint, BCA’s Gertken and Ntonifor see artificial intelligence boosting real (inflation-adjusted) bond yields, as a result of faster economic growth. Commodities and real estate could benefit, too. “People will scramble to buy land with their newfound riches, only to discover that it is the one thing that AI cannot produce more of,” the analysts conclude. Even possible “massive deflation” might raise real bond yields, as central banks struggle to increase demand to match rising output, they add.

The great fear is that AI will displace workers, causing significant unemployment. However, historically, technological innovations that increase productivity ultimately have led to rising real wages, Deutsche Bank’s team writes.

BCA cites a recent paper contending that 10% of tasks could be done via AI, affecting 80% of the U.S. workforce, with law, education, information technology, and management consulting most affected. To be sure, that would be painful for those put out of work, but DB posits that artificial intelligence’s widespread adoption might make for better and happier lives.

That is, if humanity survives the transition to superintelligent AI, the chances of which BCA puts at 50-50 by the middle of the century. They note that luminaries, including Tesla’s (TSLA) Elon Musk and

Apple

(AAPL) co-founder Steve Wozniak, have signed a letter calling for a six-month pause in AI research to develop better safety protocols.

Cybersecurity already is a $188 billion global industry, and companies that add AI safety to their security products will provide an opportunity for investors. There’s always a bull market somewhere, even if the end of humanity looms.

Pataisymai ir patobulinimai

A BCA Research report on the implications of artificial intelligence was written by Peter Berezin. An earlier version of this column incorrectly attributed it to Matt Gertken and Chester Ntonifor.

Rašykite Randallas W. Forsythas [apsaugotas el. paštu]