Let's Talk Generative AI — Simply Put

Let's Talk Generative AI — Simply Put

We have already seen a lot of videos and posts on Generative AI. So why am I talking about it again?

Well, I’m trying to explain it in — simple words and with examples from our own networking world.

What is Generative AI?

In plain terms — Generative AI is a type of AI that creates somethingunique.

It doesn’t just look at data and say “this is spam” or “that’s not spam”. It actually creates — a paragraph, an image, even config suggestions — based on what it has learned before.

So, it’s not just predicting.. it’s generating something new ..

How it Works

Gen AI has something called a neural network working at the core. Think of it like a version of our brain — lots of layers,neurons, each doing its own job.

• First layer takes your input (like a prompt or a word)

• Middle layer called hidden layer process the data and look for patterns

• Final layer gives you the output

Each "neuron" inside this network does a small calculation to decide what it should pass on to the next layer. These calculations are based on things like weight, bias, and activation function.

Let me go through each :-

Say your input has values like x1, x2, x3…

1. Weights — these decide how important each input is.
   Like, if x2 is more important attribute than x1, it gets more weight.

2. Bias — even if everything is zero, sometimes a neuron still wants to speak up. Bias makes that possible.

3. Activation Function — this decides whether a neuron should “activate” or stay quiet. Basically, it’s a filter.

The formula looks like:

z = w1*x1 + w2*x2 + ... + wn*xn + b output = activation(z)

This result moves to the next layer, and the process continues.

How Does It Learn?

The AI model learns over time through something called training.

Here's what happens:

• First, you give it input

• It gives you output

• You compare it with the actual answer

• If it’s wrong, it makes changes (by adjusting weights and bias)

  This error is measured using a cost function, which measure the discrepancy between the predicted and actual outputs.

• This process keeps repeating — and the model keeps improving

The way it fixes itself is called backpropagation .This process involves propagating the error, computed as the difference between predicted output and actual output, backward through the network.

And the method it uses to improve is called gradient descent.It is an optimization technique that allows each weight to be adjusted by calculating its degree of change.

Nothing fancy — just means the model takes small steps in the right direction till it gets better.

Let’s Bring It Into Networking

This is where it gets interesting for us.

Say you’re managing a large network — multiple sites, lots of routers, switches, logs , lot of support tickets.

Now imagine Gen AI trained on your own environment — logs, past tickets, config files, SNMP data, etc.

Here’s what it can do.

Example: Automatic RCA Report with Gen AI

Problem: You see packet loss in Region X. Alerts come in.
Old Way: You log in, check logs, dig through past incidents, prepare RCA. Hours gone.

With Gen AI:
You just type:

And it replies:

Straight to the point.

⏱️ Time saved: Big
📄 Consistency: High
📚 Backed by actual data: Yes

There are so many other case too , like config generator ,capacity manangment ,audit repot etc.

Why It Works for Us

• It handles logs, tickets, config — all the messy stuff

• It finds patterns we may miss

• It can explain things, not just detect

• It keeps getting better with feedback

Final Words

Generative AI is like a team member that doesn’t sleep. You give it data, it gives you solutions.
You ask for help, it explains. It’s not perfect — but it’s fast, useful, and always learning.So don’t think of it as a threat.
Use it as a tool — to save time, reduce noise, and work smarter.