Days in the Life of Mo

When you think about your brain, you probably don’t picture it as a rival to your laptop. It doesn’t have a power button, doesn’t crash mid-Zoom call, and it’s definitely never asked you to “install updates and restart.” But somewhere in a lab right now, engineers are teaching silicon to think, or at least to pretend to.

Welcome to neuromorphic computing: the strange, sci-fi-sounding world where computer chips start mimicking the human brain.

So… What Even Is Neuromorphic Computing?

Think of your brain as the original high-efficiency processor. It runs on about 20 watts of power (that’s less than a dim light bulb), yet it can recognize faces, learn new skills, feel emotions, and still remind you of that embarrassing thing you said in 2016. Computers, on the other hand, need enormous energy and time to perform even a fraction of that kind of processing.

Neuromorphic computing aims to change that by redesigning computer chips to work more like our brains, using “neurons” and “synapses” made of transistors instead of cells. Instead of processing data linearly, like your regular CPU, these chips process information in parallel, asynchronously, and with memory right next to computation… just like your brain’s neural network.

IBM’s TrueNorth chip was one of the first public leaps into this space. It had over a million artificial neurons and 256 million synapses, and could perform complex visual tasks while consuming just 70 milliwatts of power. Compare that to the energy-hungry GPUs used for AI today, and you start to see why neuromorphic computing feels like the next frontier.

Why Should We Care?

Because our current computers, as powerful as they are, still suck at thinking like humans.

Sure, ChatGPT can write you a poem about spaghetti in the style of Shakespeare, but behind the scenes, it’s running on massive data centers that consume megawatts of power. According to the University of Massachusetts Amherst, training one large AI model can emit as much carbon as five cars over their lifetimes. That’s… a lot of Teslas.

Neuromorphic chips could drastically reduce that footprint. They don’t just compute faster; they compute smarter. Imagine AI systems that can learn continuously, process sensory input in real-time, and adapt to new information on the fly, without needing to retrain on petabytes of data. That’s the dream.

The Brainy Copycat Wars

Right now, big names like Intel, IBM, and BrainChip are in the race to create chips that “think” more like us. Intel’s Loihi 2, for instance, is designed to simulate biological neurons and even supports spiking neural networks, models where data moves like electrical impulses through the brain.

But here’s the kicker: even if we teach machines to mimic our neurons perfectly, does that mean they understand?

Can a chip “feel” confusion? Can it daydream? Or is it just doing what it’s told, faster and with fewer existential crises?

These are the kinds of questions that make neuromorphic computing both fascinating and slightly unsettling. Because if we give computers the architecture of our minds… what stops them from developing the habits of them too? (Like overthinking, for example. Or forgetting what they walked into the room for.)

Where This Could Go

The practical uses are huge:

• Robotics: Machines that can sense, react, and learn like animals.
• Healthcare: Brain implants that process and respond in real-time to neural activity.
• AI at the edge: Tiny devices running advanced AI without needing cloud power.

In 2023, researchers at the University of Sydney even used neuromorphic chips to simulate parts of the human retina, potentially helping develop better prosthetic vision systems.

It’s innovation with a heartbeat, or at least, the blueprint of one.

A Thought to Leave You With

We’ve spent decades teaching computers to compute. Now, we’re trying to teach them to think. Maybe the next question isn’t how smart machines can become, but whether they’ll ever learn to forget, to feel, or to find meaning in the noise like we do.

Until then, I’ll keep overworking my human brain and underestimating how much RAM my laptop really needs.

---

Like what you just read?
Stick around for more curious, casually chaotic takes from your friendly neighborhood developer trying to make sense of where tech meets the human brain, literally.