I do like going further and further down the rabbit hole.
When I first became interested in artificial intelligence years ago, I immediately had a long list of questions.
What exactly is it? How does it think? Does it think at all? And is it intelligent in the same way we humans are?
So, in the second part of this series, let’s take a closer look at what artificial intelligence actually is.
Did you know that Hoover is actually a brand name?
The thing itself, of course, is a vacuum cleaner. But Hoover became so well known in Britain that the name eventually became closely associated with the product itself.
It even became a verb.
In Britain, it is completely normal to say:
“I need to hoover the carpet.”
Something rather similar is happening with artificial intelligence.
For many people, ChatGPT was the first AI-powered system they really interacted with directly.
It was no longer some invisible technology working quietly in the background. It was something you could open, ask questions, have a conversation with and turn to for help.
And from there, it is very easy to arrive at:
ChatGPT = AI.
But that is not quite how it works.
ChatGPT is not artificial intelligence itself
Hoover is just one of many vacuum cleaner brands.
In the same way, ChatGPT is just one of many systems that use artificial intelligence. And the same is true of Gemini and Claude.
They are different AI-powered systems that we, as users, can interact with directly.
Artificial intelligence is a much broader field.
It can be used in systems that recognise objects in images. It can recommend products in an online shop. It can look for unusual patterns in financial data. It can turn speech into text. It can help a car recognise road signs. Or it can allow us to have a conversation using natural language. ChatGPT, Gemini and Claude are simply some of the names we have become very familiar with.
So what is AI?
Put very simply, artificial intelligence is an umbrella term for technologies that enable machines to carry out certain tasks that would traditionally require human intelligence.
That might include recognising patterns, working with language, making predictions, solving problems or supporting decisions.
And there are many different kinds of AI systems within that broad category.
Systems such as ChatGPT made AI far more visible because, suddenly, we could actually talk to it.
But different forms of artificial intelligence had already been working quietly around us long before we started typing questions into chat windows.
Think about spam filters in your email. Recommendations from Netflix or YouTube. Banks detecting unusual transactions that might indicate fraud. Or your phone recognising faces in your photographs.
We were already using AI. We just didn’t necessarily think, “I’m using artificial intelligence right now.”
Why does this matter for a business?
Because when someone says:
“We need AI.” They still haven’t really said what they need.
It is a bit like saying:
“We need technology.” All right. But what for?
To answer customer questions? To help customers choose between products? To handle incoming phone calls? To analyse information? To reduce repetitive admin? Very different problems may need very different solutions.
That is why, at NLS, our first question is still not:
“Where can we put AI?”
It is:
“What problem are we actually trying to solve?”
Only then does it make sense to decide whether artificial intelligence is needed at all — and, if it is, what kind.
Because ChatGPT is not AI itself.
Just as Hoover is not the whole world of vacuum cleaners. It is simply one very familiar name within it. 😄