Boomy's AI is making music way faster than humans, but it hasn't written any hits — yet

Boomy is lowering the barrier to entry into music production. But what happens to streaming sites like Spotify and traditional music labels if algorithms are the hitmakers?

A photo of an audio mixer

Boomy is lowering the barrier to entry for people who want to get into the music business.

Photo: James Kovin via Unsplash

Bay Area-based Boomy has no bands, no stars, no hit producers. Despite that, the startup is churning out tens of thousands of new tracks a day. To date, Boomy users have collectively created more than 3.2 million songs. "We are technically the biggest record label in the world," said Boomy CEO Alex Mitchell in a recent conversation with Protocol.

Boomy's secret: Each song is produced with the help of AI, with users simply picking a genre, and algorithms creating a full instrumental track that can then be manually rearranged and fine-tuned in a matter of minutes.

As interesting as Boomy is as a product, its implications for the music industry are even more intriguing. What does it mean if an AI is creating music faster than humans ever could? What impact do Boomy releases have on the catalogs of music-streaming services like Spotify? How will it affect traditional record labels and musicians reliant on music licensing revenues? And who on earth is going to listen to all this AI-generated music?

In some ways, Boomy's story is the typical tale of technology lowering barriers of entry. "There's a lot of people left out of music-making," Mitchell said. Music production tools can be expensive, and turning ideas into tracks can be time-consuming and require a lot of knowledge, even with digital tools. With Boomy, all one needs is a smartphone and an internet connection.

"Anyone in the world can make music," Mitchell said. Some of the people who have embraced the service include yoga instructors looking to create original soundtracks for their classes, and parents finding new ways to get creative with their kids. (Even the intro music for Protocol's Source Code podcast was created with Boomy.)

But Boomy isn't just an AI production tool. The startup also functions as a kind of one-stop shop for people looking to release their tracks, and has established relationships with both streaming services like Spotify and music rights entities like ASCAP. "We are literally another indie record label or distributor," Mitchell said.

An indie label that could quickly become a major player in the world of recorded music, that is. Boomy's website proudly displays the number of tracks composed with the help of its AI in real time. At the time of writing, that number was 3,208,559, amounting to 3.57% of all of the world's recorded music, according to the company's calculations. And the pace of creation is increasing quickly: One million of those songs were created in the last 40 days alone.

During our conversation, Mitchell admitted that these numbers do come with a few asterisks. Boomy's AI may have created 3.2 million tracks, but the number of those actually deemed worthy to be saved by Boomy's users is closer to 800,000. Furthermore, users have to save five tracks before they can release them, and the number of releases is only now approaching 100,000.

Still, that's a lot of music, especially considering that Spotify's total catalog consists of just 70 million tracks. Thus far, Boomy's flood of releases hasn't caused any issues with digital music services, but Mitchell acknowledged that the music industry wasn't totally prepared for AI. "Some of these systems and processes have to change," he said.

One reason that Boomy hasn't made huge waves in the industry yet is that most of its releases have stayed below the surface. "We don't have any hits," Mitchell admitted. "We don't have anything close to hits." Still, some of its tracks do perform well enough on Spotify, and even get picked up by the service's algorithmic recommendations for Discover Weekly playlists.

That's when Boomy users suddenly see a bump in earnings. The company lets people keep 80% of the revenue generated by their tracks. And while a few extra dollars from Spotify may be nice, that money really is starting to make a difference for video creators, as it allows them to unlock an additional revenue stream with their clips. "We've seen a lot of growth from TikTok and YouTube," Mitchell said.

The company now plans to improve the way people can fine-tune Boomy tracks, and also add support for additional genres in the coming months. In addition, the company wants to double down on the social media angle, and perhaps even release dedicated apps for creators.

This may ultimately result in even more questions about the role of humans in the creative process. When making a song is as easy as taking a photo on your phone (and involves at least as much algorithmic processing), then what does that mean for the people clicking the buttons? "Are they artists?" Mitchell asked. "Are they musicians? Are they social media users?"

At its current pace, Boomy is likely going to produce millions of additional tracks before we have to come up with answers.

Protocol | Workplace

Pay audits catch your salary mistakes. Here's how to conduct one.

It’s not unlawful to pay people differently. You just have to be able to justify the difference.

Pay audits reveal the truth about how you’re paying employees.

Illustration: Christopher T. Fong/Protocol

This story is part of our Salary Series, where we take a deep dive into the world of pay: how it's set, how it's changing and what's next. Read the rest of the series.

In 2015, Marc Benioff famously signed off on a salary review of every employee at Salesforce on the urging of then-Chief People Officer Cindy Robbins and another senior woman executive, Leyla Seka. They suspected that women employees at the company were being paid less than men for the same work.

Keep Reading Show less
Michelle Ma

Michelle Ma (@himichellema) is a reporter at Protocol, where she writes about management, leadership and workplace issues in tech. Previously, she was a news editor of live journalism and special coverage for The Wall Street Journal. Prior to that, she worked as a staff writer at Wirecutter. She can be reached at mma@protocol.com.

The fintech developers who made mobile banking as routine as texting or online shopping aren't done. The next frontier for innovation is open banking – fintech builders are enabling consumers to be at the center of where and how their data is used to provide the services they want and need.

Most people don't even realize they're using open banking services today. If they connected their investment and banking accounts in a personal financial management solution or app, they're using open banking. Perhaps they've seen ads about how they can improve their credit score by uploading pay stubs or utility records to that same app – this is also powered by open banking.

Keep Reading Show less
Bob Schukai
Bob Schukai is Executive Vice President of Technology Development, New Digital Infrastructure & Fintech at Mastercard, where he leads the technical design, execution and support of innovative open banking and fintech solutions, as well as next generation technologies to support global payment and data capabilities. Prior to Mastercard, Schukai’s work focused on cognitive computing, financial technology, blockchain, user experience and digital identity. He is also a member of the Institute for Electrical and Electronics Engineers.
Protocol | China

Chinese ed-tech firms’ poignant pivots

Beijing’s tutoring ban has forced ed tech and private tuition companies to explore new opportunities, from clothing to coffee to agriculture.

Some Chinese online tutoring firms are pivoting away from education. Others continue offering classes, but a different kind.

Photo: Liu Ying/Xinhua via Getty Images

Management at China’s leading tutoring and ed-tech firms has been racking brains in an effort to pivot away from the once-lucrative but now moribund K-9 tutoring business. Pivoting has become a necessity ever since Beijing delivered a devastating blow to the private tutoring industry this past summer by banning many types of after-school tutoring outright.

The Wall Street Journal in November reported that several major Chinese tutoring and ed-tech companies were in discussions with China's government to resume K-9 tutoring, under the condition they run their businesses as nonprofits. But some companies have decided to sever their K-9 operations altogether, exploring completely different businesses: agriculture ecommerce, garment-making and even coffee houses. Others will stay in the business of teaching but place their bets on professional education and “well-rounded education”(素质教育), which is not schoolwork-oriented and instead involves extracurricular activities such as arts, sports, science, technology, engineering and civics.

Keep Reading Show less
Shen Lu

Shen Lu is a reporter with Protocol | China. Her writing has appeared in Foreign Policy, The New York Times and POLITICO, among other publications. She can be reached at shenlu@protocol.com, or via Twitter @shenlulushen.

Protocol | Workplace

Calendly thinks it can save you from group meeting scheduling hell

Add another tool to your arsenal for that 15-person, multi-time zone meeting you need to schedule.

Calendly now offers meeting polls.

Image: Calendly

Scheduling a one-on-one meeting over email requires its own multi-email song and dance. But scheduling a meeting for seven people over email is a full-blown nightmare. Add in multiple time zones and incomplete email responses, and you’re deep in a distressingly long email thread. So far, scheduling app Calendly has tackled one-on-one scenarios: The host sends a Calendly link and the invitee chooses the time slot that works for them. Group meetings were still a hassle, despite a few features allowing for round robins or multihost meetings. With the company’s Thursday launch of meeting polls, Calendly joins tools like Doodle and When2meet in solving group scheduling nightmares.

Srinivas Somayajula, Calendly's head of Product Operations, hopes that new and existing users will recognize Calendly as a tool for both the one-on-one use case and complex group scheduling. “We've got the capabilities in the toolset to support either of those extremes and everything in the middle,” Somayajula said.

Keep Reading Show less
Lizzy Lawrence

Lizzy Lawrence ( @LizzyLaw_) is a reporter at Protocol, covering tools and productivity in the workplace. She's a recent graduate of the University of Michigan, where she studied sociology and international studies. She served as editor in chief of The Michigan Daily, her school's independent newspaper. She's based in D.C., and can be reached at llawrence@protocol.com.

Protocol | Workplace

CTO to CEO: The case for putting the tech expert in charge

Parag Agrawal is one of the few tech industry CTOs to nab the top job. But the tides may be shifting.

Parag Agrawal’s appointment to Twitter's CEO seat is already alerting a new generation of CTOs that the top job may not be so out of reach.

Photo: Twitter

Parag Agrawal’s ascension to CEO of Twitter is notable for a few reasons. For one, at 37, he’s now the youngest CEO of an S&P 500 company, beating out Mark Zuckerberg. For another, his path to the top as a CTO-turned-CEO is still relatively rare in the corporate world.

His leap suggests that CEO succession trends may be shifting, as technology increasingly takes the center stage in business and strategy decisions not just for tech companies, but for the business world more broadly.

Keep Reading Show less
Michelle Ma

Michelle Ma (@himichellema) is a reporter at Protocol, where she writes about management, leadership and workplace issues in tech. Previously, she was a news editor of live journalism and special coverage for The Wall Street Journal. Prior to that, she worked as a staff writer at Wirecutter. She can be reached at mma@protocol.com.

Latest Stories