by

LESSON

AI 058. Examples of AI paintings or music?

listen to the answer

ANSWER

Artificial Intelligence (AI) has made notable strides in the fields of art and music, creating works that have intrigued and sometimes even astonished human audiences. Here are some examples where AI has excelled in generating paintings and music:

Paintings and Visual Art:

  • “Edmond de Belamy”: This portrait, created by a collective called Obvious using a Generative Adversarial Network (GAN), was sold at Christie’s auction for $432,500 in 2018. The artwork features a fictional character and is part of a series that explores the interface between art and AI.
  • “The Next Rembrandt”: A project that combined data from Rembrandt’s paintings with deep learning algorithms to create a new artwork, mimicking the style of the Dutch master. The result was a portrait that closely resembles Rembrandt’s work, demonstrating how AI can replicate the style of historical artists.
  • Google’s DeepDream: Initially created to help understand how neural networks perceive and interpret images, DeepDream generates surreal, dream-like images by enhancing patterns in existing photos. The results have been showcased for their artistic and hallucinogenic qualities.

Music:

  • “Daddy’s Car”: Composed by AI from Sony’s CSL Research Laboratory, this song was created using the Flow Machines software, which analyzes music databases to produce songs in various styles. “Daddy’s Car” was crafted in the style of The Beatles, showcasing the AI’s ability to create catchy, original compositions.
  • AIVA (Artificial Intelligence Virtual Artist): AIVA is an AI composer that has been trained on thousands of pieces of classical music, learning to compose original works that have been performed by human orchestras. AIVA has been recognized as a composer by the French authors’ rights society SACEM.
  • Taryn Southern’s Album “I AM AI”: Taryn Southern’s album is one of the first to be entirely composed and produced with AI, using several AI programs, including Amper Music. The album blends pop with AI’s capabilities in composition, demonstrating the potential for AI-human collaboration in music production.

These examples illustrate AI’s growing role in creative processes, generating works that engage with human traditions in art and music while also pushing the boundaries of what’s possible with machine learning and algorithmic creativity. While AI-generated art and music may differ from human-created works in terms of intent and emotional depth, they offer fascinating insights into the evolving relationship between humans and machines in the realm of creativity.

Artificial Intelligence (AI) in the realm of art and music is not only about machines working in isolation but also about the collaboration between AI and human artists, leveraging tools that bridge the gap between human creativity and algorithmic precision. Here are enhanced examples emphasizing collaborations, tools, and an analogy to bring this concept to life:

Collaborations in Art:

  • Refik Anadol’s “Machine Hallucination”: Anadol uses machine learning algorithms to process vast datasets of images, creating mesmerizing visual experiences projected on architectural spaces. His work is a collaboration between his creative vision and the AI’s ability to generate complex, evolving patterns.
  • Mario Klingemann’s “Memories of Passersby I”: Klingemann, an artist exploring AI and creativity, uses neural networks to generate portraits in real-time, displayed on antique monitors. The artwork, a machine constantly creating unique faces, represents a partnership where AI is the brush and algorithm the artist’s hand.

Collaborations in Music:

  • Holly Herndon’s Album “PROTO”: This album features songs co-created with an AI named Spawn. Herndon and her collaborators trained Spawn on a dataset of their own singing, enabling it to contribute to the album’s vocal compositions. The project exemplifies a deep collaboration between human musicians and AI, exploring new forms of musical expression.
  • IBM’s Watson Beat: Watson Beat is an AI music composition tool that understands music theory, compositions, and emotional intents. Musicians can collaborate with Watson Beat, inputting their melodies and selecting moods, to co-create compositions that are both emotionally resonant and compositionally inventive.

Tools:

  • GANs (Generative Adversarial Networks): Used in creating visual art, GANs involve two neural networks— one generating artworks and the other evaluating them—working in tandem to produce increasingly sophisticated images.
  • OpenAI’s Jukebox: A neural network that generates music, including rudimentary singing, as raw audio in a variety of genres and artist styles. It’s a powerful tool for exploring how AI can generate complex compositions.
Read more

Quiz

What significant achievement did DeepMind's AlphaFold accomplish in the field of biology?
A) It discovered a new species of bacteria.
C) It generated digital models of DNA.
B) It accurately predicted protein folding structures.
D) It automated the entire drug discovery process.
The correct answer is B
The correct answer is B
Which AI program is known for generating new artworks that mimic historical styles?
A) TensorFlow
C) DeepDream
B) Watson Beat
D) AlphaGo
The correct answer is C
The correct answer is C
How has AI impacted the music industry, particularly in composition?
A) By limiting the range of music that can be produced.
C) By enabling the creation of new music that blends different styles and genres.
B) By replacing all musicians with robots.
D) By banning the use of electronic instruments.
The correct answer is C
The correct answer is C

Analogy

Imagine a grand orchestra, the epitome of collaborative creativity, where each instrument brings a unique sound, contributing to a harmonious and complex performance. In this orchestra, AI acts as both a new kind of instrument and a collaborator, bringing its own “sound” to the music. Just as a conductor (the human artist) guides the orchestra, integrating the traditional instruments with this new one to create something beautiful and unprecedented, artists and AI systems work together, blending human creativity with algorithmic innovation to produce art and music that is both familiar and refreshingly new.

In this grand performance, tools like GANs and Watson Beat are like the unique properties of this new instrument—its ability to play a wide range of notes (generate diverse outputs) and to respond dynamically to the conductor’s direction (adapt to input from human collaborators). The result is a concert (a body of work) that transcends the capabilities of either the traditional orchestra or the new instrument alone, showcasing the potential of human-AI collaboration to push the boundaries of artistic and musical expression.

This approach to AI in art and music celebrates the fusion of human intuition, emotion, and creativity with the precision, scalability, and new perspectives that AI offers, heralding a new era of co-created art that enriches our cultural landscape.

Read more

Dilemmas

Authenticity and Value: Given that AI can create art and music that resonates with human audiences, should AI-generated works be valued differently from those created by humans? What constitutes authenticity in art and music when AI is involved in the creative process?
Intellectual Property and Authorship: As AI plays a larger role in creating art and music, who should hold the intellectual property rights to such works? Should the creators of the AI, the users who inputted the data, or the AI itself be recognized as the author?
Impact on Artists and Musicians: With AI capable of producing art and music at scale and potentially at lower costs, what are the implications for professional artists and musicians? How can the industry adapt to ensure that human creativity remains valued and viable?

Subscribe to our newsletter.