Understanding the Storage Capacity of the Human Brain and Advanced Data Technologies

Summary:

This video explores the storage capacity of the human brain and compares it to various data storage technologies.

The human brain, treated as a computer, can store an estimated 143 terabytes of information, equivalent to 55 million ebooks, by considering synapses as bits.
Replacing the brain with Hard Disk Drives (HDDs) would yield less storage (120 terabytes), but just the platters could store 1.2 petabytes.
Solid-State Drives (SSDs), particularly M.2 form factors like the WD Black SN850X, could store 1.6 petabytes in a brain's volume, surpassing HDD platters and allowing dynamic thought.
DNA, a biological storage method, can hold 800 megabytes per molecule. Filling a brain with cell nuclei (each containing DNA) could store 2 zettabytes, or 750 trillion ebooks.
Emerging technologies like Magnetoresistive Random Access Memory (MRAM) show promise, with prototypes storing 64 gigabits in a tiny space, potentially leading to 1.6 yottabytes in a brain's volume.
The theoretical upper limit for information storage in a given space is a black hole, which could store an astounding 10^55 terabytes in a human brain's volume.

The Human Brain's Storage Capacity [00:00:00]

The brain is capable of remembering a vast amount of information, from song lyrics to recognizing tens of thousands of words.
When viewed as a computer, the human brain's storage capacity is estimated by treating synapses as the equivalent of "bits."
- A synapse is the junction between two neurons, transmitting information through chemical or electrical signals of varying strengths.
- Scientists estimate that a single synapse can store between 4.1 to 4.6 bits of information.
With an average of 250 trillion synapses, a human brain can hold approximately 143 terabytes of data.
- This is comparable to about 55 million ebooks, where a typical ebook is 2.6 million bytes.

HDDs store data magnetically on spinning platters divided into tiny sections, each representing a bit (0 or 1).
A read-and-write head magnetizes these sections to store or retrieve data.
Comparing a human brain's volume to the physical size of standard HDDs:
- One of the highest-capacity HDDs, the Seagate Exos Mozaic 3+, stores 36 terabytes of data in a standard metal box (147mm x 102mm x 26mm).
- If a brain's volume were filled with these complete HDDs, it could hold 120 terabytes, which is less than the biological brain.
Excluding auxiliary parts and only considering the data-storing platters:
- Packing just HDD platters into a brain's volume could store approximately 1.2 petabytes of data, about eight times more than a human brain.
- However, this configuration would lack the ability for dynamic thought or data access.

SSDs store data using floating-gate transistors, which are electrical switches that store bits by filling or emptying a gate with electrons.
A charged gate represents a 0 (transistor "off"), and a discharged gate represents a 1 (transistor "on").
SSDs offer higher data density but have a limited number of write cycles, making them less ideal for constantly updating information like a brain.
Considering M.2 form factor SSDs for high data density:
- The WD Black SN850X M.2 SSD can hold 8 terabytes in a small space (half the size of an ID card).
- Filling a human brain's volume with these SSDs could store 1.6 petabytes of data.
- This is equivalent to 600 million ebooks and allows for dynamic data access and thought, surpassing HDD platter density by 30%.

DNA stores genetic information using four nucleotide bases (adenine, cytosine, guanine, thymine).
Each "rung" on the DNA ladder has four possible base pair combinations, equivalent to two bits of information.
A single human DNA molecule, with about 3.2 billion rungs, stores 800 megabytes of information.
This is packed into a nucleus roughly 10 microns in diameter.
Theoretically, filling a brain's volume with cell nuclei containing DNA could store 2 zettabytes of data, equivalent to 750 trillion ebooks.

MRAMs store data magnetically using two magnetic layers separated by an insulator.
One layer has fixed magnetization, while the other's magnetization changes to represent a 0 or 1.
Bits are swapped by manipulating electron spin with electric currents, avoiding HDD limitations and allowing for high data densities.
While largely in the R&D phase, a Kioxia prototype MRAM stored 64 gigabits in an area smaller than a grain of sand.
Filling a brain with this cutting-edge MRAM technology could store 1.6 yottabytes of data (1.6 trillion trillion bytes).
This is approximately 1,000 times more ebooks than DNA storage and nearly 9 times the entire world's projected data generation in 2025.

This method involves mapping one bit of information onto a single atom.
In 2018, University of Alberta scientists demonstrated storing bits using hydrogen atoms placed on silicon.
An atom's presence in a specific spot represents a 1, and its absence a 0.
This achieved a data density 47 times higher than the example SSD.
If a brain were filled with this hydrogen atom technology, it could store 75 petabytes of data (30 billion ebooks).
This lower number compared to MRAM is due to the auxiliary parts (like the silicon bed) required for the technology to function.

Black holes represent the theoretical upper limit for information storage in a given space, as packing more information would cause an object to collapse into one.
Building on the work of Jacob Bekenstein and Stephen Hawking, the information density of a black hole is calculated based on its 2D "surface" (event horizon), not its 3D interior.
If a human brain's volume were to collapse into a black hole, it could theoretically store an astonishing 10^55 terabytes of information.
- This is 10 million trillion trillion trillion trillion terabytes.
- At the projected 2025 global data generation rate, it would take 5.5 x 10^43 years to generate this amount of information, a time frame extending into the universe's "Black Hole Era."
The concept of gravity-based information storage is not purely theoretical; Gia Dvali proposed an artificial neural network that uses gravity for communication between artificial synapses, achieving black hole-like information packing without actually being a black hole.