AI and art collide on this engineering course, which puts human creativity in the primary place

Title, in fact:

Art and generative AI

What did the concept have prompted the course?

I see many students who consider artificial intelligence to be human, just because they write essays, make complex mathematics or answer questions. AI can imitate human behavior, but there is no such thing as a sensible examination of the world. This separation inspired the course and was shaped by the ideas of the German philosopher of the twentieth century Martin Heidegger. His work shows how we’re deeply connected and present on the earth. We find meaning through motion, care and relationships. The human creativity and championship come from this intuitive reference to the world. The modern AI simulates intelligence by processing symbols and patterns without understanding or care.

In this course we reject the illusion that machines completely master all the things and put the expression of the scholars in the primary place. We appreciate uncertainty, mistakes and imperfections as essential for the creative process.

This vision extends beyond the classroom. In the tutorial 12 months 2025-26, the course will include a brand new community learning cooperation with the bogus groups of Atlanta. Local artists will work with me to integrate artistic practice and AI.

The course builds on my class, art and geometry from 2018, which I work with local artists. The course examined Picasso's cubism, which represented reality from several perspectives than broken. It also examined Einstein's theory of relativity, the concept time and space usually are not absolutely and clear, but a part of the identical fabric.

What does the course explore?

We start researching the primary mathematical model of a neuron that Percepron. Then we study them Hopfield networkWhich imitates how our brain can remember a song when you only take heed to just a few notes by filling out the remaining. Next we have a look at ourselves Hintons Boltzmann machineA generative model that can even imagine and create recent, similar songs. Finally we study today deep neural networks And TransformersKi models that imitate how the brain learns to acknowledge pictures, language or text. Transformers are particularly suitable for understanding sentences and conversations, and so they power a technologies corresponding to chatt.

In addition to AI, we integrate artistic practice into the courses. This approach extends the perspectives of the scholars to science and engineering through the lens of an artist. The first offer of the course in spring 2025 was with the teachings with Mark LeibertAn artist and professor of practice at Georgia Tech. His specialist knowledge is in Art, AI and digital technologies. He taught students of the fundamentals of assorted artistic media, including charcoal drawing and oil painting. The students used these principles to create art ethically and creatively with AI. They critically examined the source of coaching data and ensured that their work respected authorship and originality.

The students also learn to record brain activity with electroencephalography – EEG – headsets. Through AI models you’ll learn to remodel neural signals into music, pictures and storytelling. This work inspired performances through which dancers improvised in response to A-generated music.

Why is that this course relevant now?

AI entered our lives so quickly that many individuals don’t fully understand how it really works, why it really works when it fails or what their mission is.

When creating this course, the goal is to strengthen the scholars by filling this gap. Whether you’re recent to the AI ​​or not, the goal is to make your inner algorithms clearly, accessible and honest. We deal with what these tools actually do and the way they will go unsuitable.

We place the scholars and their creativity first. We reject the illusion of an ideal machine, but we provoke the AI ​​algorithm to confuse and hallucinate when it creates inaccurate or nonsensical reactions. To do that, we deliberately use a small data record, reduce the model size or limit the training. In these incorrect AI states, the scholars appear as conscious fellow artists. The students are the dearth of algorithm that takes back control of the creative process. Your creations don’t obey AI, but newly recent. The artwork is saved from automation.

What is a critical lesson from the course?

The students learn to acknowledge the boundaries of the AI ​​and use their failures to regain creative authorship. The murals is just not produced by AI, but it surely is reinterpreted by students.

Students learn that Chatbot queries have environmental costs because large AI models use loads of electricity. You avoid unnecessary iterations when designing input requests or using AI. This helps to scale back carbon emissions.

What will the course prepare for the scholars?

The course prepares the scholars how artists think. Through abstraction and imagination, you gain trust to tackle the technical challenges of the twenty first century. This includes the protection of the environment, the structure of resilient cities and the development of health.

The students also recognize that AI has transitional and scientific applications, but the moral implementation is decisive. Understanding the kind and quality of coaching data that AI uses is crucial. Without them, AI systems risk having risked or incorrect predictions.