News

Creative AI

05 August 2022

Can AI be creative, really?

 

We need to talk about AI.

 

There seem to be two schools of thought around Artificial Intelligence.

 

Some industries are yet to join the conversation, seeing true AI as an event that may or may not happen in the future.

 

But others are using AI right now to transform their companies, whether that’s through Data insight recognition – Palantir style, or using software as a service (SAAS), like Google and Tesla.

 

You may have seen Google recently fire an employee, essentially for sharing company secrets. He claims Google’s Chat bot, LaMDA, may have a level of consciousness, and if you’ve read the leaked transcript, there are parts that make the hairs on the back of your neck stand up.

 

Google argues that just because their chatbot is very good at pretending to be conscious doesn’t mean that it is. That’s an argument ex-employee Lemoine found hard to stomach. Because wouldn’t that kind of thinking make it impossible to ever identify a conscious machine?

 

The fact that this is a major news story, talked about in water cooler conversations up and down the world, really highlights where we now are in the march towards true AGI.

 

“Artificial general intelligence (AGI) is the ability of an intelligent agent to understand or learn any intellectual task that a human being can.”

 

Just over 10 years ago your phone was considered smart if it could send an email, or browse the web. Now chat bots are talking about feelings of loneliness and violation, cars are increasingly navigating more and more complex road situations, and other forms of AI are giving the creative industries pause for thought.

 

The AI Artist.

 

No doubt you’ve heard about Open AI’s DALL-E software, capable of taking complex sentences and translating them into visuals.

 

DALL-E 2’s interpretation of “A photo of an astronaut riding a horse.”

 

 

 

Not only does DALL-E have access to millions of images, it has the ability to understand what those images and the subjects within them are, and more importantly, how different subjects can relate to each other.

 

This allows it to manipulate pixels, “Photoshop them”, to suit any brief.

 

Here’s a “A head of broccoli complaining about the weather”

 

 

 

And “Spice up your life, oil painting”

 

 

And “A bowl of soup that is a portal to another dimension as digital art”

 

 

How about “Portrait of a megalodon talking on the phone, comic book style”

 

 

 

 

Users can quickly edit their own photos too — replacing a dog on the sofa with a cat, adding a vase of flowers to a table, changing the walls in a room to pink, or changing the style of picture, from cartoon to photographic and from photographic to impressionist painting.

 

DALL-E will use, what’s hard to not define as “imagination” to recreate photographic pictures of Atlantis, or faces of people who don’t exist in the real world — though it still has a lot of work to do on human faces and hands.

 

Like any artist, it’ll explore the options and give you a selection of creations to choose from.

 

Is it going to take jobs from creative industries?

 

Now this is the real meat you’ve been waiting for. The reason you’re here.

 

But really, that’s the wrong way to look at it.

 

DALL-E isn’t yet able to handle the detail and constant collaboration required to make a fully-fledged campaign — so in that sense, not yet.

 

But that’s not to say creative industries should ignore technologies like DALL-E.

 

There’s room to work with these kind of technologies in the future, in ways that can help enhance our creativity and speed up our workflow.

 

For instance, all the images above were generated in about 2 seconds. Which makes DALL-E a pretty unstoppable tool for brainstorming, finding references and exploring big ideas.

 

AI is something everyone in almost every industry needs to be clued up on — not just the basic fundamentals, but knowing how its use can fine tune our work processes, increase productivity, and increase output.

 

Will robots take over in the future?

 

Whether you believe that’s likely or unlikely, right now, what looks certain is that there’s going to be an abundance of tools coming out of AI projects applicable to all types of industries.

 

Look out for them. And find new ways to work with them.

 

 

Want to learn more about DALL-E, check out Open AI’s great explainer video here.

 

 

Written by Alex Hamilton.