Pssst…Before we get into the nuts and bolts of the episode, we wanted to let you know we have a new facebook group for this podcast! 🤗

Search for Design and Prosper Podcast Community or click here.

---

We have some concerns about AI –  and so should you.

In part 1 of this series we covered all the things we LOVE about AI and how it can help elevate design businesses – go check it out here.

We are really excited about the possibilities for graphic designers, but is there a dark side to such advanced, powerful technology? In part 2 we are flipping the switch and looking at what we’re not so excited about.

As professional designers, we need to be mindful of its potential downsides and use it responsibly.

There’s huge unknown territory when it comes to AI and as with anything new, it’s natural to fear the unknown. It’s not just us that feel this way — the creators of OpenAI have said they have fears too.

Sam Altman, CEO of OpenAI said in a recent interview:

“I think people should be happy that we’re a little bit scared of this… I think if I said I were not you should either not trust me or be unhappy I’m in this job.” And, “There is a set of very bad outcomes. One thing I’m particularly worried about is that these models could be used for large scale disinformation. I am worried that these systems now that they are getting better at writing computer code could be used for offensive cyber attacks.”

So let’s break down some of the concerns we have when it comes to using AI as a designer:

QUALITY OF AI GENERATED VISUALS

It’s true that we can ask AI generators to create anything we want, like a cat in a boat wearing a rainbow sweater drinking tea. But would the result actually be any good?

We’ve seen a lot of AI artwork out there is overly complex and has questionable aesthetics. And when we’ve had our own play with the technology, it certainly leaves a lot to be desired.

AI solutions remind us of when Photoshop first got layers with opacity (yes, we were there) and designers around the globe went gang-busters with layered effects. Was it good? Not really. Did the aesthetic style last? No.

We love a good trend (see our podcast episode on designing with trends here), but only if it is appropriate and on-brief. Sure, AI can create the wildest things, but as a credible designer, it’s essential to always come back to your client’s goals and the objectives of the brief.

We need to be careful not to over-rely on AI and make sure we’re not using it just for the sake of using it. It’s a powerful tool, but it’s not a replacement for human creativity and critical thinking. We still need to use our own skills and judgment to create designs that truly resonate with our clients and their audiences.

QUALITY OF AI GENERATED WRITING

AI is said to be creating original, generative content, but is it really producing relevant and exciting new thoughts that are actually groundbreaking?

When it comes to writing, AI tends to create overly complex, repetitive solutions that feel robotic. If you don’t finesse the input and be extremely strategic with the prompts you use, it lacks personality and misses brand voice completely.

You really need to carefully consider what you ask it to do and keep on refining it. For example ‘make it more conversational’, ‘condense it’, ‘include subheadings’, ‘write it for a younger audience’….the list goes on. We’ve even tried ‘make it more loving’, to which we then had to ask it to ‘make it less cheesy’. It still has a long way to go.

AI definitely provides a good starting point, although it leaves us wondering if it takes just as long to fix what it spits out as it would be to write from scratch. It feels a bit like a lump of clay you need to mould into shape.

Bottom line is, if you put rubbish input into AI, you’ll get rubbish out. It’s really only as good as what you input, and it takes considered thought and discernment to do this.

ETHICAL CONSIDERATIONS FOR USING AI

As a community, we seriously need to consider the ethical implications of AI and where it is taking source material from. AI takes references from many different sources to create something new, but who created it originally?

It feels like the wild, wild west when it comes to legally appropriating other people’s work and the ethical dilemma that comes with that.

This seems especially problematic for AI art and visual generators – it is ethically ambiguous. Where are these apps scraping their source material from?

The legal cases are already starting. We have heard Getty Images is suing the creators of AI art software creators, Stability AI for allegedly scraping its content and breaching copyright. No doubt this will be the first case of many.

Adobe is addressing the issue of where they are sourcing their AI content from — here’s what they had to say:

“Adobe’s first model in our Firefly family of creative generative AI models is trained on Adobe Stock images, openly licensed content, and public domain content where copyright has expired.“

Adobe is also in the process of implementing a way for creators to be compensated if their work is used to make derivative works. They stated:

“We are developing a compensation model for Adobe Stock contributors, and we’ll share the details of this model when Firefly exits beta.“

That is encouraging, however until these waters get less muddied, we strongly recommend not using AI visuals for any published work for clients or yourself. It’s fantastic for brainstorming and conceptual purposes, but leave it at that for now.

COMMUNICATION BIAS

We’re concerned about AI and its potential biases, such as racial or gender bias. Currently, AI is based on a massive amount of data that has been fed to it, and this data reflects what people have created and shared pre-2021. Who says the masses are right?

Biases exist in humans, so it’s likely that AI-created works will also have biases and lack diversity. If it creates something that perpetuates harmful stereotypes or biases, challenge it! Ask it to regenerate the response, as AI is learning via our interactions with it.

WHO OWNS THE COPYRIGHT FOR AI-GENERATED WORKS?

Let’s say you create a logo using AI technology for a client. It may surprise you that this is likely to be owned by the platform you created it on – not you, and not your client.

Copyright and ownership are currently very grey areas in the AI world, and we suspect it’s shaping up to be a copyright sh*t show.

You might be wondering if copyright and intellectual property rights will even matter in the future. With new generations repurposing content all the time, it feels like the rules are unravelling a bit. But we’ve seen similar disruptions in the past with things like downloadable music, and the world tends to adjust and regulate itself over time.

This is what Adobe says about content created using AI (and no doubt other platforms will do something similar):

“Firefly will automatically attach a tag in the embedded Content Credentials to make AI-generated art and content easily distinguishable from work created without generative AI.“
Adobe

So this is interesting! Just as AI is clever at creating, AI will also be clever at embedding tags that will reveal how the work has been generated.

ACCURACY

We need to be mindful of the misinformation that may be produced through AI generators. The model will confidently state things that are facts that are completely wrong.

Here’s what ChatGPT says when you land on the main page:

• May occasionally generate incorrect information
• May occasionally produce harmful instructions or biased content
• Limited knowledge of world and events after 2021

People have shared (and will continue to share) information that is incorrect, inaccurate and even harmful. And because AI is scraping that information as source material, it’s going to spit out content based on those so-called ‘facts’.

---

So how are you feeling about it? We’d be so fascinated to hear your thoughts about AI and how you plan on using it in your design business! Send us a DM on Instagram and let’s chat.

AI is definitely going to take time for it to develop to its full potential, so stay informed, stay curious and stay in the loop as technology continues to evolve.

Much love,
Kris & Donna xo

P.S We used ChatGPT to convert our podcast transcript into this blog post! It took quite A LOT of finessing.

For full transcript click here

Wired video mentioned during episode