I want to walk you through a real, peer-reviewed, published-in-Nature study, because the headline alone sounds like science fiction and it is not.
In 2017, a team of researchers at Harvard — Seth Shipman, Jeff Nivala, Jeffrey Macklis, and George Church — encoded a digital movie into the genome of living E. coli bacteria using CRISPR. Then they sequenced the bacteria's DNA and read the movie back out.
Let me break down exactly what that means and why it should stop you in your tracks.
---
First: What Is DNA, As a Storage Medium?
Before we get to the movie, let's talk about what DNA actually is from a pure information standpoint.
DNA is a molecule that stores instructions using four chemical bases: adenine, thymine, cytosine, and guanine — A, T, C, G. That's a four-letter alphabet. Every cell in your body carries about 3 billion of these base pairs. And the information density is extraordinary: one gram of DNA can theoretically store around 215 million gigabytes of data.
For comparison: all the data humanity has ever created — every photo, video, document, tweet, database, and server farm on earth — could fit in a few kilograms of DNA. A few kilograms.
DNA also doesn't degrade fast. As of 2021, researchers sequenced DNA from mammoth teeth frozen in Siberian permafrost for over a million years — the oldest DNA ever sequenced and authenticated. Your hard drive won't last ten. Your flash drive won't last twenty. DNA, stored correctly, lasts geological timescales.
---
What the Researchers Actually Did
The study title is: "CRISPR-Cas encoding of a digital movie into the genomes of a population of living bacteria." Published in Nature, July 2017.
Here is what they encoded: Eadweard Muybridge's 1878 "Sallie Gardner at a Gallop" — the short photographic sequence of a horse running that is widely considered the first motion picture ever created. The very first movie ever made. They encoded it into DNA using CRISPR.
The CRISPR-Cas system — the same gene-editing tool that's being used to develop treatments for genetic diseases — has a natural mechanism that lets bacteria acquire short sequences of foreign DNA and integrate them into their own genome. The researchers harnessed that mechanism to write in the pixel values of each frame of the film as DNA sequences, injected frame by frame into the bacterial genome over time.
Then they sequenced the bacteria's DNA, retrieved the encoded sequences in order, and reconstructed the movie.
It worked. The images came back. The movie played.
---
Why "Living Bacteria" Is the Key Detail
Most DNA data storage research uses synthesized DNA — lab-produced strands assembled outside of any living organism. That's already impressive, but this study went further.
This was written directly into living, replicating bacterial cells.
That means when the bacteria divided — when one cell became two, two became four, four became eight — the encoded data replicated along with it. The movie wasn't just stored in DNA. It was stored in a self-copying biological system.
Every time that bacterium divided, it made another copy of Muybridge's horse.
That is a fundamentally different category of storage than anything we've built with silicon and metal.
---
Why This Is "Holy Crap" Important
Let me spell out the implications, because they go far beyond "neat science project."
1. The storage density problem is solved in principle.
The world generates more data every year than we have reliable long-term storage for. Tape degrades. Drives fail. Cloud infrastructure consumes enormous amounts of energy and physical space. DNA solves all three problems simultaneously. The Harvard group demonstrated the write-read mechanism works. The engineering to scale it is a different problem — but a solvable one.
2. It's self-replicating.
Silicon storage is passive. You write to it, it sits there, it eventually fails. Biological storage replicates itself as long as the organism is alive. You don't need a data center. You need a petri dish.
3. CRISPR is the write head.
What this paper proved is that CRISPR isn't just a gene editor. It's a tool for writing arbitrary information into living genomes. That reframes what CRISPR is — it's a biological pencil that can write anything, not just edits to broken genes.
4. The implications for medicine are significant.
If you can write data into a living genome and read it back out accurately, you can encode information into cells and retrieve it later. Think about what that means for recording biological events as they happen — a cell that logs what happened to it, in its own DNA, in sequence, over time. The same paper describes experiments encoding information that changed over time into bacteria, giving them a kind of biological memory.
5. This was published in 2017.
That's the part that gets me. This paper is not new. The proof of concept that you can write a movie into DNA using CRISPR and read it back out has been sitting in Nature for years. The implications have been there the whole time. The engineering has been moving forward since then. Where it is now, what's been built on top of it — that's the question worth asking.
---
The Choice of Film Is Not an Accident
I want to come back to Muybridge's horse for a second.
In 1878, Eadweard Muybridge captured a horse mid-gallop using a series of still cameras. The resulting sequence — a few dozen frames showing a horse's legs leaving the ground — was the first time motion had ever been captured and replayed as moving images. It was the first film.
The Harvard team chose to encode the first movie ever made into living DNA. That was a deliberate choice. They were making a point about what DNA storage represents: not just a new medium, but a new category. The first film wasn't stored on hard drives or cloud servers. It went into something alive.
---
What Comes Next
DNA data storage is an active and rapidly developing research field. The challenges that remain are speed and cost — writing to DNA is currently slow and expensive compared to electronic storage. But the trajectory is the same one we've seen in every storage technology: costs drop, speeds increase, and what seems exotic becomes infrastructure.
The concept is proven. The physics works. The chemistry works. The question now is purely engineering.
We are in the early days of something that may eventually redefine what we mean by a hard drive.
---
Sources
- • Shipman, S.L., Nivala, J., Macklis, J.D., Church, G.M. — CRISPR-Cas encoding of a digital movie into the genomes of a population of living bacteria. Nature, 547(7663), 345–349. 2017.
- • Erlich Y, Zielinski D. "DNA Fountain enables a robust and efficient storage architecture." Science. 2017;355(6328):950–954. (215 petabytes — ~215 million gigabytes — of theoretical storage density per gram of DNA)
- • van der Valk T, Pečnerová P, Díez-Del-Molino D, et al. "Million-year-old DNA sheds light on the genomic history of mammoths." Nature. 2021;591:265–269. (oldest authenticated DNA sequenced, from mammoth teeth over 1 million years old)
— Dr. Scott