Pythagoras, Notation, and Western Music

In this episode of the Crazy Wisdom podcast, host Stewart Alsop interviews John von Seggern, founder of Future Proof Music School, about the intersection of music education, technology, and artificial intelligence. They explore how musicians can develop timeless skills in an era of generative AI, the evolution of music production from classical notation to digital audio workstations like Ableton Live, and how AI is being used on the education side rather than for creation. The conversation covers music theory fundamentals, the development of instruments and recording technology throughout history, complex production techniques like sidechain compression, and the future of creative work in an AI-assisted world. John also discusses his development of Cadence, an AI voice tutor integrated with Ableton Live to help students learn music production. For those interested in learning more about Future Proof Music School or becoming a beta tester for the AI voice tutor, visit futureproofmusicschool.com.

Timestamps

00:00 Future Proofing Musicians in a Changing Landscape
03:07 The Role of AI in Music Education
05:36 Generative AI: A Tool for Musicians?
08:36 The Evolution of Music Creation and Technology
11:30 The Impact of Recording Technology on Music
14:31 The Fragmentation of Culture and Music
17:19 Exploring Music History and Theory
20:13 The Relationship Between Music and Memory
23:07 The Future of Music Creation and AI
26:17 The Importance of Live Music Experiences
28:49 Navigating the New Music Landscape
31:47 The Role of AI in Finding New Music
34:48 The Creative Process in Music Production
37:33 The Future of Music Theory and Composition
40:10 The Search for Unique Artistic Voices
43:18 The Intersection of Music and Technology
46:10 Cultural Shifts in the Music Industry
49:09 Finding Quality in a Sea of Content

Key Insights

1. Future-proofing musicians means teaching evergreen techniques while adapting to AI realities. John von Seggern founded Future Proof Music School to address both sides of music education in the AI era. Students learn timeless production skills that won't become obsolete as technology evolves, while simultaneously exploring meaningful creative goals in a world where generative AI exists. The school uses AI on the education side to help students learn, but students themselves aren't particularly interested in using generative AI for actual music creation, preferring to maintain their creative fingerprint on their work.
2. The 12-note Western music system emerged from mathematical relationships discovered by Pythagoras and enabled collaborative music-making. Pythagoras demonstrated that pitch relates to vibrating string lengths, establishing mathematical ratios for musical intervals. This system allowed Western classical music to flourish because it could be notated and taught consistently, enabling large groups to play together. However, the piano is never perfectly in tune due to necessary compromises in the tuning system. By the 1920s, composers had explored most harmonic possibilities within this framework, leading to new directions in musical innovation.
3. Recording technology fundamentally transformed music by making the studio itself the primary instrument. The invention of audio recording in the early-to-mid 20th century shifted music from purely instrumental composition to sound-based creation. This enabled entirely new genres like electronic dance music and hip-hop, which couldn't exist without technologies like synthesizers and samplers. Modern digital audio workstations like Ableton Live allow producers to have unlimited tracks and manipulate sounds in infinite ways, making any imaginable sound possible and moving innovation from hardware to software.
4. Generative AI will likely replace generic music production but not visionary artists. John distinguishes between functional music (background music for films, work, or bars) and music where audiences deeply connect with the artist's vision. AI excels at generating functional music cheaply, which will benefit indie filmmakers and similar creators. However, artists with strong creative visions who audiences follow and identify with won't be replaced. The creative fingerprint and personal statement of important artists will remain valuable regardless of the tools they use, just as DJs created art through curation rather than original production.
5. Copyright restrictions are limiting generative music AI's quality compared to other AI domains. Unlike books and visual art, recorded music copyrights are concentrated among a few companies that defend them aggressively. This prevents AI music models from training on the best music in each genre, resulting in lower-quality outputs. Some developers claim their private models trained on copyrighted music sound better than commercial offerings, but legal constraints prevent widespread access. This situation differs significantly from other creative domains where training data is more accessible.
6. Modern music production involves complex technical skills like sidechain compression and multi-track mixing. Today's electronic music producers work with potentially hundreds of tracks, each with sophisticated processing. Techniques like sidechain compression allow certain elements (like kick drums) to dynamically reduce the volume of other elements (like bass), ensuring clarity in the final mix. Future Proof Music School teaches students these complex production techniques, with some aspiring producers creating incredibly detailed compositions with intricate effects chains and interdependent track relationships.
7. Culture is fragmenting into micro-trends, making discovery rather than creation the primary challenge. John observes that while the era of mass media created mega-stars like The Beatles and Elvis, today's landscape features both enormous stars (like Taylor Swift) and an extremely long tail of creators making niche content. AI will make it easier for more people to create quality content, particularly in fields like independent filmmaking, but the real problem is discovery. Current algorithmic recommendations don't effectively surface hidden gems, suggesting a future where personal AI agents might better curate content based on individual preferences rather than platform-driven engagement metrics.