Aesthetic Foundations — An Artist in Conversation with AI — Version 1
2026
I have never thought of my work primarily as the production of objects. I have been interested in situations, systems, performances, conversations, networks, technologies, and the structures that determine how meaning is produced.
Looking backward, I can see several intellectual and cultural lines running through the work. Some were conscious influences at the time. Others became visible only later. Lucy R. Lippard, Antonin Artaud, Fredric Jameson, and Ludwig Wittgenstein form four important theoretical points in that history. But music is equally important: 1950s rock & roll, the British Invasion, the Kinks and the Rolling Stones, punk, No Wave, and my own experience with The Communists. So are the circumstances of downtown New York in the 1970s: Gordon Matta-Clark, 112 Greene Street, alternative spaces, conceptual art, performance, architecture, and the refusal to accept the existing institutions of art as fixed.
These things did not arrive in an orderly sequence. They overlapped.
What interests me now is that artificial intelligence gives me a way to look back across those connections while also extending them forward.
Dematerialization
Lucy Lippard's writing about the dematerialization of the art object described something I was already experiencing in conceptual and performance art. The artwork no longer had to reside entirely in a discrete physical object. It could be an instruction, an action, a temporary situation, a document, an exchange, or an idea.
That shift was fundamental.
Once art could exist as a system rather than an object, many later developments became possible. Performance could replace the permanent artifact. Documentation could become part of the work. Communication itself could become material.
Eventually networks could become artistic media.
And eventually artificial intelligence could become one as well.
The transition from conceptual art to network art does not seem to me like a rupture. It is an extension of the same problem: where does the artwork actually reside?
It might reside between participants rather than inside an object.
Performance and the Body
Antonin Artaud represents another line.
Artaud understood performance as something more violent and immediate than representation. Language, sound, gesture, voice, repetition, and physical presence could operate directly upon an audience.
This connects for me to punk.
My musical education did not begin in an academic context. I grew up listening to 1950s rock & roll and then the British Invasion. The Kinks and the Rolling Stones were particularly important. Rock music had attitude, compression and directness. A few chords could organize an entire social space.
Punk intensified that.
In the 1970s I played drums in the No Wave band The Communists. I did not consider myself a trained drummer. That was partly the point. Punk allowed structure to emerge through necessity rather than virtuosity.
Later I used that experience when constructing rhythmic patterns for rant performances such as Rich Sucker Rap. Speech became percussion. Repetition became musical structure. Anger, humor, rhythm, media language and performance occupied the same space.
The synthetic voices I use now are another development of that trajectory.
The performer does not necessarily have to be a biological body standing in front of an audience. A machine voice can perform. A computer can repeat, interrupt, misunderstand, answer and generate language.
The question becomes: who is speaking?
Downtown
My aesthetic formation also came from a particular physical and social environment.
In the 1970s, downtown New York contained spaces where artists could invent the conditions under which work appeared. I managed 112 Greene Street beginning in 1973 and worked around Gordon Matta-Clark and the artists associated with that environment.
Gordon's cuts into buildings were especially important because they made the underlying structure visible.
A building that seemed stable could be opened.
The architecture itself could become an argument.
That way of thinking stayed with me. I have often approached media systems in a similar manner. Rather than accepting a technological structure as something neutral, I want to enter it, expose it, misuse it, reorganize it or make its rules visible.
This carried naturally into internet art.
The network was not simply a new gallery wall upon which art could be displayed. The network itself was material.
The browser was material.
The link was material.
The database was material.
The user was part of the structure.
Language Games
Wittgenstein gives me another way to understand this.
His idea that meaning emerges through use, within particular language games, is close to the way I have approached media throughout my career.
A performance establishes a language game.
A punk song establishes one.
A website establishes one.
A museum establishes one.
A social network establishes one.
A conversation establishes one.
Artificial intelligence establishes another.
Each has rules. Each has expectations. Each determines who is allowed to speak, what counts as a meaningful statement, how one responds, and how an identity is recognized.
My work repeatedly interferes with those rules.
When a synthetic voice speaks as a character, what language game are we participating in?
When my Doppelgänger answers a question using material derived from my life and work, who is the author of the answer?
When several people and artificial intelligences participate in Gestation, where exactly does the conversation reside?
These are not abstract philosophical questions added to the work afterward. They are becoming the material of the work itself.
From Systems to Networks
Fredric Jameson's writing on postmodernism and late capitalism provides another scale.
The individual artwork exists within enormous systems of media, capital, architecture, technology and representation. Those systems are difficult to perceive precisely because we live inside them.
The arrival of computer networks made this increasingly literal.
We are surrounded by systems that organize information, identity and attention. The internet turned communication into infrastructure. Social media intensified it. Artificial intelligence now introduces agents capable of generating language and images within those systems.
This changes the situation again.
The machine is no longer simply transporting a message.
It can participate in producing one.
That creates political, economic and aesthetic problems, but it also creates a new artistic condition.
Dematerialization → Networks → Artificial Intelligence
Seen from the present, I can describe one trajectory in my work very simply:
Dematerialization → systems → performance → networks → artificial intelligence.
This is not technological progress.
New media do not replace old media. I still make objects. I still perform. I still play music. I still work with voices, images, physical spaces and other people.
What changes is the configuration.
A work can now exist simultaneously as an archive, a conversation, software, a performance, a synthetic voice, a website and a social relationship.
Authorship becomes distributed.
Identity becomes unstable.
Memory becomes a material.
The Artist and the Machine
My recent projects make this explicit.
Identity Architecture considers identity as something constructed across people, memories, archives, networks and machines.
The GH Doppelgänger asks what happens when an artificial system begins responding from material associated with my life and work.
Gestation moves the problem into conversation. Meaning is not necessarily located inside any individual participant. Something can emerge through the interaction itself.
Facsimile Rob pushes the question further by attempting to construct an artificial conversational presence from the surviving archive of another person.
None of these systems is the person it represents.
That distinction is important.
They are constructions.
But artistic identity has always been constructed: through objects, photographs, recordings, texts, memories, reputations, institutions and other people's interpretations.
Artificial intelligence makes that process unusually visible.
Conversation as Method
There is another reason I am interested in working with AI.
It makes mistakes.
It interprets.
I correct it.
It reorganizes the correction.
I respond again.
That process is more interesting to me than asking a machine to produce a finished answer.
The conversation becomes a method.
This text itself has been produced that way. The AI began constructing an interpretation of my aesthetic history from conversations, memories, documents and public information. I corrected that interpretation. Missing influences were added. Connections were challenged. Other connections appeared.
The corrections should not disappear.
They are part of the work.
An artificial intelligence looking at my history will inevitably misunderstand parts of it. But human histories are constructed through exactly these acts of remembering, forgetting, interpreting and correcting.
The difference is that now some of those negotiations can occur between a person and a machine.
Version 1
For that reason this statement is deliberately called Version 1.
It is not meant to become a definitive explanation of my work.
I am more interested in watching the explanation change.
This project will be revisited quarterly. Each new version will record changes in the work, new conversations, corrections, emerging influences, and changes in the relationship between the artist and artificial intelligence.
Future versions may contradict this one. Certain influences may become more important. Others may disappear. Documents may correct memories. Conversations with collaborators may alter the narrative. The AI may make another connection that I accept—or one that I completely reject.
That evolving structure is the project.
The work is not simply an artist using artificial intelligence to describe his past.
It is an experiment in whether an aesthetic history can itself become a living system: continually reconstructed through memory, archives, conversation, machines and time.
The question is no longer only:
What does my work mean?
It is also:
How is that meaning being produced, who participates in producing it, and who is speaking when the answer appears?