Brain uploads could become possible as PCs get more powerful

Death might be old-hat soon if a university neuroscientist's predictions are correct. We'll be hosted by computer.

Artificial intelligence computer brain circuits electronics grid

Living inside a machine is "definitely a possibility," according to a British neuroscientist. Dr. Hannah Critchlow,of Cambridge University, says that if a computer could be built to recreate the 100 trillion connections in the brain, it would be possible to live within programs.

Critchlow's statement, at the Hay Festival, was reported by the Telegraph and Metro newspapers.

The Hay Festival is an annual writers' gathering in the UK. Bill Clinton once called it "the Woodstock of the mind." The Telegraph is a sponsor.

Circuit board

Critchlow says that even though the brain is complex, it works "similarly to a large circuit board" and that "scientists were beginning to understand the function of each part."

Because of that, it would be possible to eventually download consciousness, she says. I think, as others do too, that she meant to say "upload," or was misquoted.

Critchlow has a PhD in neuropsychiatry. In 2014 she was named a top 100 UK scientist by the Britain-based Science Council for her work in science communication.

Death could be old-hat

"Death could soon be a thing of the past, with old people hooked up to computers to 'download' their brains into ultra-powerful machines, where they can live on forever," the Metro says of the idea.

There are those who don't agree, obviously:

The brain is "a complex mixture of electrical, and chemical reactions and processes. We can't even define consciousness, let alone measure it, so how are we supposed to create or copy it?" Telegraph commenter Winterwarm says.

Artificial intelligence circuits

But interestingly, matters of metaphysics aside, engineers are actually on track with new artificial intelligence circuits. Critchlow's predictions might not be as far-fetched as you might at first think.

Scientists at the University of California Santa Barbara and Stony Brook University reckon they've made a breakthrough. They say that they've gotten an artificial circuit to perform the complex task of sorting through simple images—a kind of learning.

And it's doing it with chips that "borrow design points from the brain," MIT Technology Review says. They're called neuromorphic chips.

Classifying letters

The group was able to classify three letters – "z", "v" and "n" – by their images.

Big deal, you might say. But it is because they're doing it "not with digital logic circuits, but with elements that mimic, in simplified form, the neurons and synapses of biological brains," MIT explains.

Synapses are the impulse-carrying junctions between nerve cells.


The scientists have been using complementary metal-oxide-semiconductors and adjustable terminal resistive devices, called memristors. You can read more about them here in Nature.

The system even worked when each letter was stylized or inundated with visual noise.

Neural circuitry

In a method similar to how we find the correct key from a ring of similar keys, the neural circuitry was able to properly organize the simple images.

And it was accomplished not through software and heavily taxed computers, but nanocell chips that actually mimic a human brain's signals.

So Chritchlow's ideas might not be so wacky after all.

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.

Copyright © 2015 IDG Communications, Inc.

IT Salary Survey: The results are in