Share this...

By Trisha Prabhu

Hey Trish, I keep hearing about AI art but don’t know what that means. What exactly is it?

Hi there, and thank you so much for this fantastic question (what a great way to end our Ask Trish posts for 2022)! First and foremost, I wanted to validate your confusion. You’re spot-on in noting that AI-generated art is all the buzz these days – if you’re anything like me, you’ve likely seen it in articles and all over social media and heard friends and colleagues talking about it constantly. But like so many new technologies or tech-related concepts (e.g., Bitcoin, NFTs, etc.), it can be a little difficult to disentangle the buzz words from the substance, and to understand what exactly AI-generated art is all about. All that’s to say: you’re certainly not the only one wondering #what’sup.

But not to fear, because in this post, I’ll try and clear up the confusion with a brief overview of AI-generated art. In particular, I’ll briefly chat about 1) what exactly AI-generated art is, 2) why some folks are so excited about it and the opportunities it presents and 3) why those same folks and other folks also have concerns about it and the risks it may pose. Keep in mind that this is meant to be a starting point for learning about AI-generated art. I hope you use this post as a “launch pad” for further research and learning.

So, what exactly is AI-generated art? It’s exactly what it sounds like: art that is created with the assistance of artificial intelligence. Simple enough, but let’s break that down further. What’s artificial intelligence, or AI? AI is the simulation of human intelligence by machines like computers. A very popular form of AI – and often, the form of AI that’s used to help generate art or other forms of “creativity” – is machine learning. Machine learning algorithms use sample data to build a model in order to make predictions or decisions. In the case of AI-generated art, artists and computer scientists “feed” machine learning algorithms thousands of images of a certain type of art or style (the sample data) in an effort to “teach” the algorithms about that art or style. The algorithms then use what they have “learned” to create their own patterns or styles, many of which push the boundaries or art. (To learn more about the technology behind AI-generated art, check out this fantastic resource.)

A big part of the reason AI-generated art has seen so much buzz and interest lately is that there’s been a huge influx of AI art generators available to the public online. Consider, for instance, DALLE.E 2, an AI system that produces “realistic image and art” from text prompts entered by a human user. (Try it out for yourself – it’s a fantastic way to get a sense of the possibilities of AI-generated art!) Another key reason that AI-generated art is all the rage right now is how “good” the art is getting: it’s increasingly common that humans can’t tell the difference between art created by humans and art created by machines. As mentioned, machine learning algorithms are also pioneering their own creative styles of art based on what they’ve learned – a truly fascinating feat.

That brings us to the opportunities associated with AI-generated art. Simply put, the biggest opportunity is the way AI has the potential to fundamentally disrupt and transform art. AI can do everything from create new types of art to facilitate a 3D art exhibition that enables users (through additional technologies, like AR and VR) to touch and interact with the art. (Think about how cool that sounds!) AI-generated art can also be used as a medium for brainstorming for artists and designers; one can imagine them using generators to first explore (and push the boundaries of) their ideas, after which they themselves can create in new and interesting ways.

But…and yes, there always seems to be a but!…AI-generated art also poses risks and lots of challenging questions. For instance, AI-generated art can make us question what art really is. Is an algorithm that produces novel patterns or style an “artist,” and are the works it produces “art”? On the one hand, some of the things machines are producing are incredible, but on the other hand, calling these works “art” may minimize human artistry and creativity. (And indeed, it’s unclear whether we can really say algorithms are being “creative” when they produce AI-generated art.) There’s also the question of ownership. Who owns AI-generated art? The algorithm? The engineers that built the algorithm? The individual who inputted the text to produce the art (and likely strategically used that text to refine what was produced)? Again, these questions are not easily answered, which is a challenge given the rise of things like NFTs. (To learn about NFTs, check out this Ask Trish post.) And finally, there are also concerns about the way AI-generated art can facilitate some of the Internet’s worst harms, including cyberharassment and exploitation of youth. You can probably imagine numerous ways AI-generated art could be leveraged to abuse others or to say mean, hurtful things. There are few that argue that some applications of AI-generated art should not be allowed, and that some lines will absolutely need to be drawn.

I hope this week’s post offered some helpful information and insight on an up-and-coming technology that’s likely to only grow in popularity/that you will definitely see more of. Like so many things that are Internet/tech-related, AI-generated art simultaneously presents incredible opportunities and risks – which is all the more reason to be educated about it.

As mentioned, this is the last Ask Trish post of 2022! This year, we took on topics ranging from the metaverse to the pros and cons of filters to online diet culture. Hopefully, you came away with new knowledge, valuable insights, and most importantly, a sense of confidence as you take on and navigate the Internet. Thank you all so much for such an incredible year of learning together, building community, and spreading positivity online. I’m so grateful for your incredible questions, your engagement, and your commitment to this initiative.

I can’t wait to see you in 2023! I’ll be back on Tuesday, January 3rd – when we’ll explore another topic – potentially, a topic you’ve been wondering about! With that said, before I say goodbye for the rest of the year, I want to encourage you to share your Internet-related questions or thoughts – whether similar to this post or completely different – here. Remember, anything is fair game, so don’t hesitate to share whatever it is you’ve been thinking about. I can’t wait to hear from you and share some #AskTrish advice!

See you all in the New Year! Until then, wishing you and your families a safe, happy holiday season and a wonderful start to 2023,

Trish

@asktrish

Found yourself seeing article after article (and image after image!) of AI-generated art, and wondering what the buzz is all about? This week, get the scoop on asktrish! Link in bio ⬆️ #ai

♬ original sound – Brainrot

Share this...