Generative AI Debuts on Broadway

Generative AI tools have been slowly coming online for about half a decade, but the debut of ChatGPT in November 2022 is generally considered to be the moment that the world started paying attention, largely because that’s when anyone could play with OpenAI’s version of the tech for free (and soon you won’t even need to log in to use it, which will likely drive even more daily use). Many of us spent 2023 experimenting with various generative AI tools, being amazed by what was possible, sharing the results with friends and colleagues, and trying to figure out how these tools might be used in the workplace. Now, in 2024, many companies are bringing corporate AI tools into their official IT stack and investing in team accounts for their employees, including in the offices of Broadway. I sat down with one of our industry’s most tech-forward luminaries, Tim Kashani, CEO of Apples and Oranges Studios, to ask him about how he sees generative AI being used in the workplace, what he’s excited about, and how companies can keep training data safe.

Peter Yagecic: Tim, thanks for taking the time to discuss moving generative AI tools from a parlor trick to part of the official tech stack within our industry. First, can you talk about the moment you realized these tools were more than just a curiosity and deserved to be taken seriously in the workplace?

Tim Kashani: The “aha” moments for me came three times; initially, back in 1986 during my computer science studies, when the potential of AI first opened up so many opportunities. Back then, we were predicting it would be about 10 years to full adoption. Oh well, close enough. The next watershed moment came in 2016 with DeepMind’s AlphaGo victory over a world champion Go player, showcasing AI’s ability to tackle complex, creative problem-solving, solidifying my belief in its transformative potential for the arts and beyond. The day ChatGPT was released into the wild, I knew everything would change forever.

Peter Yagecic: In collaboration with The Shubert Organization, you’ve actually held a number of workshops to introduce generative AI tools to a wider swath of Broadway leadership. Why did you feel this was something you wanted to do?

Tim Kashani: We felt it crucial to empower our industry, as quickly as possible, to ensure an ethical adoption of AI within our community. We’re focusing on leveraging these tools to aid shows in development and in reaching broader audiences, while also ensuring artists are at the forefront of this conversation. It’s about democratizing technology to inspire responsible innovation and inclusivity in storytelling. We started the day acknowledging the risks and ethical implications we’ll all face in the coming months.

Peter Yagecic: What are some of the challenges you’ve experienced getting people to fully understand the potential of generative AI in our industry?

Tim Kashani: Surprisingly, we’ve encountered very little resistance. Once people move past the initial fear and truly dive in, they quickly recognize both the immense benefits and the inherent challenges of the technology. 

Peter Yagecic: Are there any early success stories from your own experimentation you can share or things in progress now that you hope will pay off?

Tim Kashani: Absolutely! Our experience has shown that AI, when combined with human expertise, creates an incredibly potent team. Most excitingly, one of our attendees, who initially had minimal knowledge of AI, was inspired during our course. By the next session, she had launched a passion project that was live and attracting users. This quick transformation from learner to innovator underscores the transformative potential of AI in empowering individuals and reshaping our collaborative landscape. 

Peter Yagecic: Let’s talk about data security. Many people have heard that any data provided to public generative AI tools can be used to train these models and could even end up in answers to queries from strangers. What can organizations do to make sure their data stays proprietary?

Tim Kashani: Addressing data security is more straightforward than many realize. With a paid account for a Large Language Model, like ChatGPT, starting at around $20 a month (per user), organizations can opt out of having their data used to train these models. This option is available with some free models as well. The key is to be proactive. While it requires some initial setup, ensuring data remains proprietary isn’t difficult. 

Peter Yagecic: In your opinion, are there particular departments within an organization that are the best candidates to “go first” in using generative AI to enhance their productivity?

Tim Kashani: In my view, it’s imperative that every department begins evaluating and learning about generative AI immediately. The landscape is evolving at an unprecedented pace, with billions being invested weekly. I can’t stress enough that the cornerstone of ethically adopting generative AI lies in understanding it thoroughly first.

Peter Yagecic: Compensation for artists whose work is used in the training of these systems is a big topic in the generative AI discussion, and each week, we see headlines detailing a new development on this front. What should our industry be mindful of as conversations in this area move forward?

Tim Kashani: You’ve certainly saved the most challenging question for last, and it’s one I contemplate daily. In our non-profit, Apples and Oranges Arts, we empower artists with our mission of “Taking the Starving out of Artist.” Therefore, I am in support of the artists and the “open letter” referenced in the article. While we operate within a commercial industry, I believe we have a more critical role in leading humanity forward through the power of storytelling. I might seem equivocal because there isn’t a simple answer to this pressing issue. Given the unprecedented pace of technological advancements in our lifetime, especially with AI now not just supporting but actively engaging in the creative process, it’s imperative we continually address this topic. We’re striving to involve as many artists as possible in these conversations immediately, recognizing this is an existential question that touches on the essence of what it means to be human. So, in leaving you with a broad philosophical perspective, I hope to underscore the significance of this ongoing dialogue in shaping our collective future. Cue the dancing cats!

Peter Yagecic: Tim, thank you so much for this conversation and for the work you’re doing to demystify generative AI for our industry!

If you’d like to join the ongoing conversation around how generative AI will impact the business of Broadway, you can register for our upcoming webinar AI + The Law 2024 on Wednesday, April 10th at noon Eastern.