AI set to change the way we search the web

I didn’t write that introduction to this article. I asked Google’s AI chatbot Gemini to do it for me by giving it a quick description of what this article is about – how it is predicted we will soon be snubbing traditional search engines in favour of AI chatbots.

Just like encyclopaedias became redundant once we had search engines at our fingertips, search engines like Google start to also feel old-fashioned and tedious once you spend a few hours with an AI chatbot like Gemini. You ask it a question and it gives you an answer – there’s no trawling through websites to find the info you need.

There are definitely some risks that come with the speed and convenience these tools give us though. Spend a bit of time with one, and it won’t take long to find the chatbot has given you bad information.

The theme of this year’s World Consumer Rights Day (on 15 March) is ‘Fair and responsible AI for consumers’, with a focus on AI chatbots being used for search. Consumer organisations around the world are urging caution when using chatbots and calling on industry leaders and governments to ensure consumers get trustworthy and bias-free information.

To talk about how we should approach this new technology in our own lives, I spoke to Albert Bifet, who is the director of the Artificial Intelligence Institute at the University of Waikato.

Kia ora Albert, can you tell me – do you see these AI language tools replacing traditional search soon?

Yes, it’s coming. How we do search is going to change. Instead of doing a query in Google we are going to be asking these large language models and then they’re going to retrieve that information. So, it’s going to change how we access the web and it’s the first time I think that Google is in trouble.

With AI set to make getting information a lot faster, what things do we need to be aware of?

A big issue at the moment is they make mistakes. They make what we call hallucinations. And that’s important because if people are asking about important things, like their health, then we don’t want these systems making a mistake. So yeah, it’s coming, but I think it still has to do some improvements. There are so many people working on these, so many companies, so many researchers. So, I can imagine all of this is going to be here very fast.

Are there any other things we need to be aware of as we start to use these tools?

We need to remember to still do some critical thinking. With traditional search we need to still do some thinking, right? We can see the websites it’s pulling the information from, so we can do some thinking about the origins of the results. I think the main issue of these AI technologies is it’s making life really, really easy. And sometimes that’s not good, right? Because we still need to think and exercise our brain. If we don’t think too much then we are not using our brain and that’s going to be a bad thing in the future. It’s important that we are the ones using this as our tools and not letting it be the decision-maker.

Is there anything else we should be aware of?

There will be a lot of scams, so the first piece of advice is to be careful. This technology is still very new, so I think don’t trust anything and double-check just in case. I can imagine that scammers will be trying to use it.

It sounds like a risky thing to start using, do you think we should be yet?

I think it’s good technology to start playing with. But I think we need to be careful because there are the hallucinations, and we still don’t know how the decisions around its recommendations are being made. Imagine if we’re searching for the best restaurant in Auckland – we don’t know why it’s recommending those restaurants and it’s different to the ones you might get when you use Google. So, my main advice at the moment is it’s good to have a play with because this is coming. But for the moment it’s just not mature enough. It will happen fast though. So maybe at the end of this year the technology will be good enough.

Consumer organisations around the globe have spent the last few weeks conducting an experiment by asking AI chatbots a series of questions ahead of World Consumer Rights Day. We will update this article once the results have been assessed and revealed by Consumers International.