Is Generative AI Real? Pun Intended

First of all, generative AI is a specific category of artificial intelligence techniques and models designed to create new content. It gets trained on existing data and then “learns” how to generate new text, images, and audio by using patterns and relationships from the training data.

The goal of generative AI is to mimic human creativity. Let’s say that again… mimic human creativity. And you thought creativity was only a human function. The key is that the new creative content made by AI must make sense and appear authentic.

Here’s where it gets more technical. Generative AI uses a technology called generative adversarial networks (GANs) which is made up of two neural networks (neural as in a nerve network… again, and you thought that was only for humans). These two neural networks consist of a generator and a discriminator.

The generator creates the content while the discriminator evaluates what’s been created and compares that to real data. These two processes compete with each with the generator trying to create content that is not distinguishable from real data (in other words it appears real) while the discriminator keeps upgrading itself to identify real from generated. As time goes on, the generator becomes better at producing realistic content, and that’s where the success of artificial intelligence lies.

Are you still with me? Because there’s more.

Generative AI uses autoregressive models like GPT (Generative Pre-trained Transformer) that predict the next element in a sequence so it can generate content that is coherent and relevant. These models use a sequence based on previous elements and generate content one element at a time. Okay, enough of the technologically confusing.

So where can generative AI be used in a practical sense?

On websites, if it is generating text, it can be used in chatbots. It can also create images that closely resemble real photographs. It can even create original music compositions – no more having to find something usable that doesn’t infringe on a copyright. And that’s just a portion of what it can do in the sense of how travel and tourism may use generative AI. It is also being used in other fields such as pharmaceutical research and gaming content.

As time goes on, it is expected that generative AI will continue to take the lead in impacting a wide range of automated tasks that involve creativity. Ethically we have to ask, is this all just a little bit too human?