In March 2023, as AI hit the headlines, Goldman Sachs released a report on the “enormous economic potential of genetic AI”. The authors explored the possibility of a “productivity explosion,” comparable to those that followed seismic technological changes such as the mass adoption of personal computers.
About 15 months later, Goldman Sachs published another paper on artificial intelligence, this time with a completely different tone. This had a crude title — “Gen AI: Too Much Spend, Too Little Benefit?” — and featured scathing reviews from executives such as Jim Covello, Goldman's head of global equity research. “AI bulls seem to just trust that the use cases will multiply as the technology evolves,” Covello said. “But 18 months after genetic AI was introduced to the world, not a single truly transformative — let alone cost-effective — application has been found.”
This skepticism has been echoed elsewhere. Daron Acemoglua prominent MIT scholar, published a paper in May arguing that artificial intelligence would lead to “much more modest productivity effects than most commentators and economists have claimed.” David Kahn, a partner at Sequoia Capital, warned in June that “we must be careful not to believe the delusion that has now spread from Silicon Valley to the rest of the country, indeed the world. This delusion says we're all going to get rich quick.”
“I worry that we're doing this hype cycle of measuring aspiration and calling it adoption,” he says Kristina McElheran, assistant professor of strategic management at the University of Toronto, who recently published a paper examining businesses' efforts to implement AI technology. “Usage is harder than ambition.”
The music industry is no exception. A recent survey of music producers conducted by Tracklib, a company that supplies artists with pre-cleaned samples, found that 75% of producers said they don't use AI to make music. Among the 25% who played with the technology, the most common use cases were to help with highly technical and decidedly unsexy processes: stem splitting (73.9%) and mastering (45.5%). (“For now, AI has shown the most promise in making existing processes — like coding — more efficient,” Covello noted in the Goldman report.) Another multi-country survey published in May by the Reuters Institute found that just 3% of humans have used AI to produce audio.
Right now, people use AI products “to do their homework or write their emails,” he says Hannah Kalert, cultural trends analyst at MIDiA Research, which recently conducted its own research on the adoption of AI technology. “But they're not interested in it as a creative solution.”
When it comes to assessing the impact of artificial intelligence—and the speed with which it would remake every aspect of society—some recalibration was probably inevitable. “There was so much excitement and promise around the launch of ChatGPT, especially since this is a technology that we talk about in pop culture and see in our movies and TV shows,” he says. Manav Raj, assistant professor of management at the Wharton School of the University of Pennsylvania, who studies business responses to technological change. “It was very easy to start thinking about how it could be really transformative.”
“Some of that excitement might have been a little frothy,” he continues. “Even if this is a really important and big technology, it takes time to see the impact of these kinds of technological changes on the markets.” This was famously true with the development of computers—in 1987, the economist Robert Solow he joked, “You can see the computer age everywhere except in productivity statistics,” a phenomenon later dubbed “the productivity paradox.”
It also takes time to settle the legal and regulatory framework governing AI technologies, which will likely affect the magnitude of their impact as well. Earlier this year, major companies sued two genAI music platforms, Suno and Udio, accusing them of copyright infringement on a massive scale. in recently filed court documents, the companies said their activities were legal under the fair use doctrine and that the major labels were simply trying to eliminate “a threat to their market share.” Similar lawsuits against AI companies have also been filed in other creative industries.
However, when McElheran surveyed manufacturing companies, few cited regulatory uncertainty as a barrier to AI use. He points out that “they may have had bigger fish to fry, like no way.” A US Census Bureau survey of businesses published in March found that 84.2% of respondents had not used AI in the previous two weeks and 80.9% of companies that did not planned to implement AI in the next six months believe that “it is not applicable to this business.”
Tracklib's research found something similar to McElheran's. Only about 10% of respondents said copyright concern was a reason they would not use AI tools. Instead, Tracklib's results showed that producers' most common objections to using AI were ethical, not legal – explanations like, “I want my art to be mine.”
“Genetic AI comes to this wall where it's so easy, it's just a push of a button,” Kahlert says. “It's a fun gimmick, but there's no real investment on the part of the user, so there's not a lot of value they actually put into the result.”
Instead, MIDiA's research found that respondents were interested in AI technology that can help them tweak tracks by adjusting the beat — a popular TikTok change that can be done without AI — and adjusting song lyrics. That interest was particularly strong among younger music fans: More than a third of 20- to 24-year-olds were interested in AI tools that could help them play to the beat, and about 20% of that age group liked the idea of can to personalize the lyrics of the songs.
Antony Demekhinco-founder of music AI company Tuney, sees a market for “creative tools” that allow “creating, editing or remixing beats and songs without the use of complex DAWs, while giving users a sense of ownership over the output.”
“Until recently,” he adds, “the addressable market for these kinds of tools has been small because the number of producers using professional production software has been limited, so early-stage technology investors don't often support such things. ”
Demekhin launched Tuney in 2020, long before the general public thought of products like ChatGPT. In the wake of that platform's boom, “investors started throwing money at it,” he recalls. At the same time, “no one knew what questions to ask. What is this trained in? Are you at legal risk? How easy would it be for Meta to reproduce this and then make it available on Instagram?”
Today, investors are much better informed, and conversations with them sound very different, says Demekhin. “Cooler heads prevail,” he continues. “Now there's going to be a whole wave of companies that make more sense because people have figured out where these technologies can be useful — and where they can't be.”