Can AI help opera companies reconnect with their communities?

I’m increasingly convinced, after two years of working on a project to track and map how AI is moving through cre­ative industries, that the path forward for opera isn’t about AI replacing any of the components of opera. Rather, a key component of the way forward will be companies seizing the chance to rebuild and own the digital infrastructure that connects them to their communities.

The lesson from the last 30 years of the dig­ital revolution is that Big Tech platforms are rapacious in inserting themselves between insti­tutions and their communities. There is potential for AI to counteract this intrusion, but only if institutions are ready to claim it.

I’ll come back to this idea in a moment. First, we need some honest accounting of where we are. As someone whose job is to scan the news of the arts world every day for ArtsJournal, the past year has been grim. Week after week, stories detail what can only be described as the whole­sale collapse of structures that supported creative industries for the past 60 years. Nonprofit, for-profit, big institution, small, individual artist — no one is immune.

Our artists and institutions are extraordinary. As the models began to fail, and particularly since the pandemic, the arts sector has engaged in introspection that has led to many positive changes.

Still, you can’t tweak yourself out of climate change. You have to adapt.

The numbers confronting the opera field are sobering. At larger companies, average operating expenses have nearly doubled in the quarter century since 1999 while the number of productions has dropped by roughly a third, according to OPERA America data. Box office income, once 41% of revenue, now covers barely 12–20%. New ticket buyers are up 15%, but these newbies are exploring, not committing, and are far less likely to follow the traditional paths to becoming subscribers and donors. The field is spending more, producing less, earning less, and relying on a donor pipeline under strain at every level. And opera isn’t alone. Versions of this run across the nonprofit arts and commercial entertainment world.

The Value Proposition

How did this come about? Here’s one answer: For almost three decades, a handful of technology platforms have done something far more destructive than competing with insti­tutions. They redefined what “value” means in the creative sector. In the old economy, the value was the product, the performance, the creative work. People paid for art and per­formances because it was intrinsically worth something. But, today, the tech platforms have shifted value away from the content itself to the traffic that content generates. A Beethoven symphony and a cat video are of identical value to the platforms’ algorithms. The embedded algorithmic values, I’d argue, are actually anti-art, elevating outrage and oddity over thoughtful creative expression.

The algorithms have largely severed relationships between artists and audience. Opera companies, orches­tras, and theaters haven’t just lost audience members over the years. They have lost the civic ecosystem — local journalism, civic identity, education funding — that made performances part of the community. The Big Tech platforms captured the space between institution and community and dictated who gets access to what and on what terms.

So what to do? I see two possible slivers of light. First, the most recent assault on our institutions may actually be the best thing that has ever happened to the case for rebuilding them. For decades, the erosion of the civic mid­dle, or middleware, as I call it, was a quiet phenomenon. The connective tissue dissolved so gradually that peo­ple didn’t notice what they’d lost, only that public life felt thinner and angrier.

The infrastructure is visible now. The closing of the Kennedy Center, the demolition of federal agencies, the defunding of public media — this is wholesale destruc­tion, not erosion. For the first time in a generation, millions of Americans are asking what institutions actually do and what will happen when they’re gone. Artists and arts leaders ought to be in the middle of these conversations, making the case that they are essential civic glue.

The second shaft of light is AI. Okay, I know. You’re tired of hearing about AI. Everyone from McKinsey to your annoying smarty-pants nephew has assured you that AI will “transform” your work, typically while trying to sell you something. But here’s the thing about transformative technologies: They don’t care whether you’re ready. The printing press didn’t wait for the Church. The internet didn’t pause while newspapers figured out a strategy. AI is not going to slow down while opera completes a strategic planning cycle.

The question isn’t whether AI reshapes this field. It is whether opera companies engage with it or not. The ques­tion is who builds the new infrastructure, and who owns it.

Future Applications

Let’s start with a visible application of AI, because these can be thrilling. The staggering cost of opera’s spectacle — sets, costumes, rigging, storage, transport — makes large-scale opera economically irrational. A major production can cost companies millions of dollars in physical infrastructure that gets used for a few performances and then sits in a ware­house somewhere in New Jersey. Generative AI could change this equation. Imagine a stage where real-time AI produces immersive environments that respond — to the singer’s voice, to dramatic arc, to the emotional temperature of the room. Such a holodeck is not a science fiction conceit; it will be available within the decade.

Using technology in this way does not diminish specta­cle; it dematerializes it. Imagine the creative possibilities to extend the art form. Computer gaming changed expec­tations of a player’s ability to interact with a story. What might some version of interactive opera look like?

But synthetic scenography, as spectacular as it might be, is still just opera doing opera differently and cheaper. The deeper opportunity, here, is what AI makes possible at the institutional level — and this is where I come back to the concept of middleware, the civic tissue that connects the art, the culture, with a community.

Think about what tech platforms did to arts institutions. They didn’t compete with them directly. They built a layer between arts institutions and audience experience online and operationalized it, deciding who got reached, what got surfaced, and what a “relationship” with a cultural organiza­tion looks like. What they were really building, underneath all the engagement metrics and algorithmic feeds, was a specific model of the connection between digital cultural content and community. That model — who responds to what, who connects to whom, what draws people in, and what leaves them cold — became the infrastructure. And institutions, desperate for visibility, handed over the raw material to build it without ever owning a piece.

Today’s AI gives institutions the tools to build a model themselves that allows for direct interaction with the art using what I like to call a “digital twin.” This would not be a CRM upgrade or a fancier mailing list, but a living, institution-owned intelligence about its relationship to its city. It could be a specialized Large Language Model trained on the specific DNA of a specific production, loaded with the script, scores, director’s concept notes, costume sketches, rehearsal videos, historical context, archive recordings, producer interviews, and a framing of the work. As I wrote in ArtsJournal, the potential patron lands on the digital twin and instead of reading a blurb, opens a dialogue.

Such a “twin” could also help companies identify which neighborhoods are culturally adjacent to opera but never reached. Which civic leaders in tech, in social justice, in education share the company’s values but have never been asked. What does the opera mean to people who’ve never bought a ticket? A well-built digital twin doesn’t just answer these questions once; it keeps answering them, updated with every interaction, every performance, every commu­nity touchpoint. The institution stops renting its audience relationship from a platform and starts owning it.

And the marketing pitch for opera shifts from “let me tell you what we’re doing,” which promptly disappears into the Sea of Infinite Messages, to “let me engage you in a con­versation about what might matter to or inspire you.” The new middleware is interaction at scale rather than mes­sages in a bottle.

The Irreplaceable Voice

This may sound theoretical, but the stakes here are not abstract. The Big Tech platforms became indispensable because they created the model before institutions under­stood what was being built. An opera company that builds its digital twin owns something genuinely scarce: a deep, reciprocal, AI-mediated map of its place in the civic life of its city. It’s no mystery why every Fortune 500 company out there is investing in assembling its own digital twin. An opera company that waits will find, again, that someone else built the infrastructure and is charging rent.

The case for opera has always been that the human voice is irreplaceable, that there is something in the embodied expression of a trained singer, on stage, in a room with other human beings, that no synthetic content can touch. I believe that. But in an age when AI generates infinite aesthetic stuff at effectively zero cost, “irreplaceable” needs to be made explicit. The value of understanding that this was made by a human, in this place, for this community, in this tradition — the provenance premium — is essential. But only if institutions claim it actively, rather than assuming everyone already understands it.

This is a clarifying moment. AI flooding the world with cheap cultural content doesn’t diminish what opera is; it illuminates it. Opera’s greatest strength has always been the same thing that makes its economics so difficult: It is stubbornly, irreducibly human. The voice, the body, the room, the risk of it — these resist automation by definition. Almost without anyone saying it out loud, that has become opera’s most valuable asset in a world awash in synthetic content.

The institutions that build their own relational infrastructure rather than renting it from platforms will look, 10 years from now, like visionaries. The ones that wait for the technology to stabilize, or for someone else to build the tools, will find themselves in a familiar position: locked out of the infrastructure that connects them to the people they exist to serve.

We really do need to learn the lessons of the last time Big Tech transformed the world. ■ 

Key Trends

A 25-year study of the opera field reveals a prevailing tension between programmatic consolidation, rising costs, and diminished return from ticket sales. The largest companies have reduced seasons by nearly half while expenses have nearly doubled. Festival companies have experienced similar spikes in budget with more moderate reductions in performances. Mid-sized companies have remained more steady but have relied on budget efficiencies. For all companies, box office revenue as a percentage of overall revenue has dropped — for many, by 50% or more.

The data below represents a constant sample group drawn from OPERA America’s Professional Opera Survey. Graphs are linearized with select years shown for simplicity. Dollar amounts shown are not adjusted for inflation. Unfortunately, sufficient data over the period was not available to compare the smallest companies in Budget Groups 4 and 5.