Restaurants and bars across the United States are increasingly finding the cost of music licensing fees untenable, prompting some to explore AI music alternatives as a way to maintain ambiance without breaking the bank. This growing conflict pits small businesses striving to create inviting environments against songwriters who depend on royalties for their livelihoods. Under U.S. copyright law, which mandates these fees, venues must obtain licenses from Performing Rights Organizations (PROs)—a system rooted in a 1917 Supreme Court decision involving composer Victor Herbert. The requirement applies whether the music is performed live, played from recordings, or streamed.
Traditionally, three main PROs: The American Society of Composers, Authors and Publishers (ASCAP), Broadcast Music, Inc. (BMI), and Society of European Stage Authors and Composers (SESAC), managed these licenses. However, the rapid rise of streaming services and the expanding value of music rights have encouraged the formation of new organizations, including Global Music Rights in 2013, AllTrack in 2017, and Pro Music Rights in 2018. This proliferation has significantly increased the financial and administrative burdens for businesses. Establishments must now potentially secure licenses from all six major PROs to avoid legal exposure, as many modern songs involve multiple songwriters affiliated with different organizations.
For restaurants and bars, these licensing costs have escalated into a significant operational concern. According to the National Restaurant Association, smaller establishments pay an average of $4,500 annually, which can represent 0.5% of their total sales, a substantial figure for businesses operating on narrow margins. Hotels have reported even more dramatic increases, with some properties seeing licensing fees jump by 200% to 400% between 2021 and 2025. Fees vary widely depending on venue size, occupancy limits, type of music use, and frequency. For instance, ASCAP’s lowest annual fee is roughly $365, but restaurants and nightclubs may pay between $1,000 and $7,000. BMI fees range from $365 to $10,000, and SESAC, a for-profit PRO, averages around $700 annually. Compounding the issue, businesses must often pay overlapping fees to multiple PROs to ensure they are covered for the songs their patrons expect to hear.
The rationale for these fees lies in the unique catalogs held by each PRO. ASCAP and BMI alone represent over 20 million and 22 million songs, respectively, and the blanket licenses they offer cover access to these vast libraries. However, businesses argue that this system forces them to pay for songs they never use and punishes them with duplication when a single track involves multiple songwriters from different organizations. Moreover, the complexity of songwriter affiliations makes it nearly impossible for a business owner to know which licenses they actually need to remain compliant.
The music industry’s financial growth has further fueled concerns. ASCAP’s revenue nearly doubled between 2010 and 2024, reaching $1.8 billion. With increased revenues from streaming and new licensing avenues, PROs have grown more aggressive in enforcement and pricing. Small businesses have criticized these organizations for opaque pricing structures and difficulty in negotiating or predicting costs. In February 2025, in response to a U.S. Copyright Office Notice of Inquiry, numerous businesses and industry stakeholders submitted public comments raising concerns about the increasing number of PROs (performing rights organizations), lack of transparency, and the burdens of compliance. While not all comments came as a single letter, the collective feedback represented widespread industry frustration.
This compliance isn’t optional. Under 17 U.S.C. § 504, businesses can face statutory damages of $750–$30,000 per work, rising to $150,000 for willful violations. ASCAP, BMI, and other PROs do pursue enforcement, recent litigation includes a March 2025 suit against 15 U.S. venues seeking up to $30,000 per work, and a Tennessee case where ASCAP won $40,000 over unauthorized playback of merely four songs. These penalties, coupled with rising baseline licensing fees, have prompted some businesses to consider dropping music altogether, potentially at the expense of customer satisfaction and revenue.
Faced with these pressures, businesses are exploring possible workarounds. Exemptions exist for establishments under 3,750 square feet using radio or TV music, but these apply only under narrow conditions and exclude streamed or live music. Hiring performers who play only original music could eliminate licensing needs, but customer preference for familiar songs makes this impractical for most venues. Commercial streaming services like Pandora CloudCover offer negotiated licenses that simplify compliance, but they too can cost $250 to $2,000 annually. Even these services don’t eliminate the need for broader licenses if songs from unaffiliated PROs are played.
This increasingly complex and costly environment has led some business owners to consider a radical alternative: replacing traditional music with AI-generated compositions. Recent technological advances have made it possible for AI systems to generate real-time, adaptive music that is not subject to traditional copyright constraints. Tools like DeepMind’s Magenta Real-time, Suno AI, and Lyria allow users to create high-quality music instantly using text prompts, environmental inputs, or even simple humming. These models are trained on extensive datasets and can produce 48 kHz stereo audio in near real-time, enabling personalized, venue-specific soundscapes without legal entanglements.
The appeal of such tools is clear. AI-generated music offers a potential escape from the burdensome web of licensing fees, especially for businesses that need continuous background music rather than specific chart-topping hits. Some AI services are even integrating with digital audio workstations, allowing business owners or musicians to customize the music further. These systems are already being used in immersive environments where music adapts dynamically to crowd energy or biometric inputs.
However, this technological promise is not without its limitations. Experts and musicians argue that while AI can mimic styles and structure, it lacks the emotional depth, storytelling, and cultural authenticity of human-created music. Surveys suggest that 82% of listeners may not distinguish AI music from human music in blind tests, but many still value the emotional connection and authenticity that comes from knowing an artist is behind a piece. AI may be well-suited to background ambiance but may not satisfy customers expecting popular or emotionally resonant songs.
Further complicating the picture are unresolved legal and ethical questions. If an AI is trained on copyrighted music, who owns the resulting work? Artists have raised concerns about unauthorized use of their material in training datasets. Industry leaders stress the need for transparency and regulation to ensure fair use and protect human creators. Without such frameworks, there’s concern that AI-generated music could flood streaming platforms, undercutting independent artists and devaluing musical craftsmanship.
Despite these uncertainties, some experts predict that AI music will become a common tool rather than a replacement. Musicians such as Paul McCartney and Holly Herndon have already integrated AI into their work, viewing it as a collaborator rather than a competitor. The future may favor hybrid models in which AI enhances human creativity, especially in settings like restaurants or hotels where music supports ambiance more than artistic expression.
The U.S. Copyright Office’s ongoing inquiry into licensing practices—prompted by mounting concerns from small businesses—has not yet resulted in reforms, but pressure is mounting. As lawmakers consider adjustments and AI tools continue to mature, the relationship between music, business, and law is poised for transformation. For now, restaurants and bars remain caught between the rising cost of compliance with traditional licensing systems and the promise, and potential pitfalls, of new technologies that could redefine how music is used in public spaces.
