The film on this page contains one image. Every shot in it is a crop of that image, a change of exposure, a length of time spent looking. Nothing appears on screen that does not come from the single frame the film opens on. The voice you hear is mine.
The image is not the painting. The painting is called The Void, and I made it as a teenager. A black field. A narrow gray band set low. A warm off-white ground beneath. It is packed away in storage while work is done on my house, and I cannot stand in front of it.
I could have waited until the house was finished. Instead I wrote down what I remembered. I gave the words to a machine I have spent months teaching, and the machine gave back images. Some I let go. One I kept, because it was the one I could answer for, the one image I was willing to put my name under.
The fields meant something when I painted them, and they mean it still. The warm ground is the world of understanding and knowledge. The black is everything we do not understand. The gray band between them was deliberate even then. It is the seam, the place where the unknown becomes known, where knowledge is created. The whole picture turns on it.1
The film opens on the image and closes on the same frame.2 Between the two appearances it tells how the image came to exist and what the choosing asked of me. I wanted something I could not reach. I described it. A machine made possibilities inside the limits my words set.3 I weighed them, chose one, and stood behind the choice. What the standing taught me feeds the next attempt. The film walks through that account in seven movements. If the closing frame reads differently from the opening frame, the film has done its work.4 If it reads the same, that is an answer too, and it is yours.
That walk is also the shape of a discipline I work in. It is called Autonomous Decision Science, and it concerns how people and machines share serious decisions. The written description I gave the machine has a name in that field. An ontology is a structured account of the known world, the parts named and the relationships among them stated plainly enough that a machine can work with what people have already come to understand.5 It is the warm field of the painting, written down.
Human-like reasoning lets a machine move across the relationships in that account and propose what no one has named yet, arrangements that may lie past the edge of what it was given. A proposal is not knowledge. The machine does not understand what it has proposed, and it cannot tell whether the thing it points to belongs in the warm field or the black. The crossing happens at the seam, and the seam is human ground. A person connected to the world weighs the proposal against everything else the world has taught them, and whatever crosses, crosses because someone judged it sound and accepted what would follow if the judgment was wrong. In this film that someone was me, and the stake was one image of a remembered painting. The stakes are usually larger. The machine did real work at nearly every step of this account. What it could not hold was the wanting at the start and the answering at the end, and the film shows both staying with me.
The painting is still in storage. When the house is finished I will stand in front of it again, and I will find out what the image on this page has done to the memory that made it. Until then, the film is the account I can give. The gray band holds it.
Stephen DeAngelis
Princeton, NJ
August 2026
About the Author
Stephen F. DeAngelis is the founder, president, and CEO of Enterra Solutions and Massive Dynamics, two companies that apply artificial intelligence and advanced mathematics to complex enterprise challenges. His work spans international relations, national security, and commercial technology, with visiting research affiliations at Princeton University, Department of Chemistry, the Computing and Computational Sciences and National Security Directorates of the Oak Ridge National Laboratory, the Software Engineering Institute at Carnegie Mellon University, and the MIT Computer Science and Artificial Intelligence Laboratory. He holds patents in autonomous decision science.
Inspirations and Endnotes
1: Kurt Gödel, "Über formal unentscheidbare Sätze der Principia Mathematica und verwandter Systeme I," 1931, <https://doi.org/10.1007/BF01700692>. The one primary mathematical source in this film's company. The theorem concerns consistent formal systems strong enough for arithmetic, and nothing wider. The paper and the painting share a single subject, a limit met from the inside, and they share nothing else. The film borrows no claim from the theorem and does not apply it to image generation, to painting, or to the authority of people over machines.
2: Douglas Hofstadter, *Gödel, Escher, Bach*, 1979. The source of the film's recursive shape, a system that climbs through another level of representation and returns to itself changed. I have reread it across three decades of work in machine reasoning. Company for the architecture, and no substitute for Gödel's paper.
3: Sol LeWitt, "Paragraphs on Conceptual Art," *Artforum*, Summer 1967, <https://www.artforum.com/features/paragraphs-on-conceptual-art-211354/>, and "Sentences on Conceptual Art," *0-9*, no. 5, 1969, <https://monoskop.org/images/3/3d/LeWitt_Sol_1969_1999_Sentences_on_Conceptual_Art.pdf>. "The idea becomes a machine that makes the art." The indefinite article matters, one device among many an idea could become. Sentence 12 adds, "For each work of art that becomes physical there are many variations that do not." An instruction bounds a field of outcomes, few become physical, and the author stays answerable for what counts as faithful. Structural company only, neither a strange loop nor evidence for anything in Gödel.
4: M. C. Escher, *Drawing Hands*, lithograph, 1948, and J. S. Bach, *Musical Offering*, 1747, above all the Canon per Tonos. The loop seen and the loop heard. Escher's two hands draw each other into existence on one sheet. Bach's canon climbs by modulation until it returns to its opening key a level higher, a strict return in counterpoint, different in kind from Escher's mutual making. The film's two held tones are drones and claim no contrapuntal kinship with the canon.
5: Ludwig Wittgenstein, *Tractatus Logico-Philosophicus*, 1921, proposition 5.6, the limits of my language mean the limits of my world. A separate philosophical source, apart from the company above, and the sentence the film's third movement circles without naming.
