A short history · The tools of music
Every era feared its new instrument.
Music has always been human — but the tools keep changing. From the bone flute to the multitrack tape to the synth to the sampler, each new instrument was doubted first and indispensable later. Here's the long view, and where today's AI tools fit in.
Before history was written
The voice and the drum
Music begins with the body: a voice humming to a child, hands clapping, feet stomping, a hollow log struck in rhythm. Every culture we know of made music before it made writing. It was not a product — it was how people mourned, celebrated, told stories and passed time together. The first instrument was a human being.
c. 40,000 years ago
The first instruments
Flutes carved from bird bone and mammoth ivory survive from the Ice Age — carefully spaced finger holes, made by people who clearly understood pitch. Drums, rattles, lyres and harps followed across every continent. Tools arrived, but the music stayed human: an instrument only speaks when someone plays it.
c. 1000 CE
Writing music down — notation
Guido of Arezzo helped invent the staff, and for the first time a song could outlive its singer. Notation was the first great music technology: it let ideas travel across centuries and borders, made complex polyphony possible, and created the role of the composer. But a page of sheet music is silent — it still takes a human to make sound.
1450s
The printing press
Printed sheet music spread songs faster than any troubadour could. For the first time, a melody written in Venice could be hummed in London within months. Music became something that could be distributed — a shift that repeats every time a new copying technology appears.
1877
Sound itself is captured — the phonograph
Edison's phonograph recorded sound onto tinfoil, and music was no longer tied to a moment or a room. The gramophone, vinyl and the microphone followed. Listening became something you could do alone, in your home. The performer, the record and the audience split apart — and the modern music industry was born around that gap.
1920s–1950s
Radio, tape and multitrack
Radio carried one performance to millions at once. Then magnetic tape changed everything: performances could be cut, spliced, layered and corrected. Les Olson-style multitracking — building a song part by part on separate tracks — turned the studio itself into an instrument. The Beatles' Sgt. Pepper could never have existed live. Music became something you construct, not just perform.
1960s–1980s
Synthesizers, drum machines and the sampler
The Moog synth, the Roland TR-808, the Fairlight and the sampler created sounds no acoustic instrument had ever made — and the 808 became the heartbeat of hip-hop, house and pop. Purists protested that machines would kill real music. Instead, humans bent the machines to new genres no one had imagined. Every one of those sounds still began with a person turning a knob or programming a pattern.
1980s–1990s
MIDI, CDs and the home studio
MIDI let keyboards, drum machines and computers talk to each other; the CD made digital audio the standard; affordable computers put a multitrack studio on a desktop. A teenager in a bedroom could do what once required a record label's budget. The tools kept shrinking and the number of people making music kept growing.
2000s
DAWs, Auto-Tune and the internet
Digital audio workstations like Pro Tools, FL Studio and Ableton Live made recording, editing and arranging infinite and reversible. Pitch correction went from a repair tool to a signature sound. Napster, iTunes, YouTube and eventually streaming rearranged the whole business again — music became abundant, and attention became the scarce thing.
2010s
Anyone can release
Laptops, cheap interfaces, free plugins and direct-to-listener platforms meant an independent artist could write, record, produce and release a song without ever signing a contract. The gatekeepers moved from labels to algorithms — and independence stopped being a compromise and became an identity.
Now
The newest instrument is AI.
Like every tool before it, AI in music is not one thing. It helps tune, clean, mix and perform. It can also generate entire songs on its own — and that is where the important line sits.
AI session players
Today's newest instrument is the same idea as the 808 or the sampler, one step further: AI that can perform. Give it your hum, your riff, your lyrics — and it can play drums, strings or backing vocals around your material, the way a bandleader directs session musicians.
Voice tools
Recording cleanup, pitch guidance and — with consent — a model of your own voice that can sing a guide take while you're away from the mic. Your voice, your lyrics, your melody: the tool sings, you author.
Assisted mixing and mastering
Software that listens to your mix and suggests levels, EQ and loudness targets. It doesn't decide what the song should feel like — it helps you get there faster, like a second pair of trained ears.
Generative engines
Systems that can produce complete songs from a text prompt. They are genuinely new — and genuinely contested, because a full song with no human authorship raises hard questions about whose music trained them and who, if anyone, owns the result. That's the line White Bunny was built around.
The tool changes. The artist doesn't.
The bone flute didn't replace the voice. The tape machine didn't replace the band. The sampler didn't replace the drummer it borrowed from — it needed one first. AI is the next step in that same story: a powerful instrument that still needs a human to point at something and say that. At White Bunny, the artist writes the song, sings the words and leads the band — AI plays in it, only with permission, and every release says so. A song made entirely by AI, with no human behind it, isn't welcome here.