Ever wondered what happens when you send an AI prompt? Here’s the story

You hit send. Your AI prompt vanishes from your screen. What happens next is invisible, but the physical journey is remarkable.

As soon as you press submit, your text turns into binary: zeroes and ones. Your computer divides this into small data packets and sends them over your home network using either Wi-Fi or Ethernet technology to reach your router.

From there, the data packets go along fibre optic cables to your ISP's regional facility. And then off they go to their final destination: an AI data centre, possibly thousands of miles away in another continent.

If your destination data centre is outside your country, things get even more interesting. The data packets will be converted into infrared light beams and transmitted through fibre optic cables laid on the ocean floor. Light travels through glass at about 125,000 miles per second.

That’s fast enough to cross the entire Atlantic Ocean and strike a data centre on the opposite coast in 40 to 80 milliseconds. You probably won’t notice it, because your brain measures the delay it perceives as 100 milliseconds, kind of like it’s instant, or so it feels.

Once your prompt arrives, it's routed to a particular server rack that has AI chips, GPUs, TPUs, and other specialised processors. The server puts together your full prompt and feeds it into the large language model.

The LLM then processes what you sent, does calculations across billions of parameters, and starts generating an answer.

But you don’t sit there until the full reply finishes computing. Instead, as the GPU outputs token by token, the server immediately wraps each chunk and sends it back toward you, right away.