Boston prepares for a symphony that weaves live AI around its players

At Boston’s Symphony Hall this week, an AI system will have a part to play in the relationship between an individual musician and an orchestra. Kent Nagano will conduct the Boston Symphony Orchestra in the world premiere of Tod Machover’s already and not yet on 22 October, with further performances on 23 and 24 October.[3]

Machover has built the work around a musical and political idea: how individual voices contend with a group, and how a damaged collective might rebuild itself. In his account of the piece, the three movements travel from confrontation through awakening to an imagined, renewed democracy. Electronics accompany the orchestra, while a specially developed live AI system becomes particularly prominent in the second and third movements.[1]

A solo line can become a larger musical space

The composer describes two different jobs for the system. During the second movement it develops variations and enhancements around solo lines, drawing attention to differences between players. In the third, it works with the orchestra’s layered sound and extends that sound into the hall. The intended progression moves from an emphasis on particular musicians towards a wider ensemble in which detail remains audible.[2]

MIT’s account adds a physical detail: additional loudspeakers will carry the extended sound beyond the stage. It also identifies the people responsible for making this happen. PhD candidate Manaswi Mishra designed and built the AI systems; fellow PhD candidate Ana Schon will handle the complex live sound mix. Composer, conductor, players, software designer and mixer therefore remain visible in the work’s chain of creative decisions.[1]

That division matters to how the piece can be understood. The score supplies a larger shape, musicians give phrases their particular character, and the electronic treatment changes what the room hears. A system that highlights a soloist and one that blends an ensemble serve different artistic purposes, even if both are described as AI. The meaningful question is what relationship the composer has made possible between those sounds.[2][1]

The premiere is still ahead as of 21 October, so these descriptions set out the creators’ intentions. How the layers sound together will be a question for the performances.[3]

A long conversation about responsive instruments

The BSO is placing the premiere within its Technology and Our Humanity programme. A 24 October conversation will pair Machover with composer George Lewis, whose Voyager software dates to the 1980s and was developed for live musical exchanges with human performers. That history gives the week a useful frame: artists have been exploring how computational systems listen, respond and participate in performance for decades.[3]

A separate research project shows where Machover intends to take the question next. The Hasso Plattner Institute lists a collaboration with MIT running from September 2026 to August 2029, led by Machover and HPI’s Walid Maalej. It will investigate AI assistance in music composition and app design, with an emphasis on human creativity. Its stated concern is that reliance on text prompts can steer people towards similar, average results.[4]

That project is distinct from the symphony’s performance system. The connection is an artistic question about the form of interaction. A performer can contribute through timing, phrasing and sound; a designer may discover an intention through successive changes. Supporting those processes requires attention to how people create, beyond accepting a verbal description of a finished result.[4][2]

Machover has also described a melody that struggles to emerge in the first movement, appears in fragments in the second and becomes central in the third. Following that thread will give the audience a way into the work’s argument through the music itself.[1]

Report an error or suggest a correction