Is AI a friend or foe? Machine Uprising or a friend to brands?

Geoffrey Hinton, the man who is widely seen as the father of Artificial Intelligence quit about a month ago, after issuing a warning about the growing dangers from the developments that were happening in this field. He is quoted to have said that some of the dangers of AI were ‘quite scary’.

It was Dr Geoffrey Hinton's pioneering research that he did on neural networks and deep learning that has laid the foundation and has paved the way for current AI systems like ChatGPT.

The neural networks that are present in artificial intelligence, or the systems in artificial intelligence, are very similar to the ones that are present in the human brain. The similarity is in the way the neural networks learn about something and process the information. And just like human beings, they help AI to learn from experience. So the AI system, just like a human being, learns as it goes along. This process is called deep learning.

According to Dr Hinton, the danger was that the chatbots could overtake the level of information that a human brain holds, very quickly. Chatbots already know a lot more and have more general knowledge than an average person. They don’t reason as well has humans. But that could change very quickly. As could a number of things!.

The thing about AI is that it changes constantly. You think you know something today and a week later, it’s almost totally outdated! And that is because, true to its name, it learns and evolves – by leaps and bounds.

Here are some characteristics of AI as we know it now – and what it means to businesses and brands –

AI is capable of passing off fake information as the truth!

We say – Now isn’t that something good for the marketing department? Won’t this be a treasure trove of creative output that marketers can do magic with in unique and most unexpected of ways?

Brands can tap into this and use it to come up with the most creative of messaging that can really go a long way in capturing audience’s attention by the sheer creativity of it all. This could be a great way to pursue some out of the box, unusual concept or fresh ideas. Could be worth it!

The most common application of AI is a Chatbot. And they seem to come with an attitude!

We say – Chatbots are good. And some personality might be interesting too! Brands can use the power of a chatbot to engage with customers better. But businesses must remember that they need to train the bot (yes you can train an AI bot) to speak the brand’s message. Businesses need to integrate bots in the whole fabric of the company so bots can pick up insights and qualitative data to converse with and engage customers.

Chatbots are great when it comes to chatting with customers. They can understand the needs and preferences of the customer, absorb all kinds of data about them and in a very engaging conversational manner find out what exactly they feel about the brand!

AI is biased

We say – Train the bias out! It depends on the data that has been used to train the AI models. The bias could have come from the ones creating the AI models as well as the data used to train the said models on.

But when the models are trained in diverse data with varied perspectives, then it is possible to remove biases or at least reduce the levels to a very large extent. By constantly monitoring the AI and checking it to see how the biases work, and correcting it, will work to reduce bias.

Like humans, Artificial Intelligence needs time and investment to train the systems that are deep in its neural networks. So instead of just treating it as a fun component that can only entertain, with the right kind of investment in training the system, brands stand to gain huge dividends.

Brands need to be open to adapting and re-defining what they and we think / thought we know. And that can truly pay off!