AI Is Ruining Art — But Saving Animals

Why I’m using AI even if it makes people mad

When you read my Substack articles, you’re getting a window into my evolving thoughts—some freshly formed, others rooted in years of experience. I don’t claim to have all the answers (even if it seems that way at times), and I invite you to engage with my ideas as just that: thoughts worth considering.

People are once again outraged with AI image generation. This time, the focus is on ChatGPT with its latest update bringing advanced image generation capabilities. But I think many people are missing something important in this debate, something that impacts our work for animals.

Just recently, I posted an AI-generated reimagining of a lovely photo I had taken in India. It’s a bit of a trend to post these Studio Ghibli-inspired generations, so I thought I’d give it a go and see if I could generate interest in my advocacy through it. Vegan fishing if you will.

The original photo, 100% real

The AI Ghibli-inspired generation

The artistic ownership debate

Since posting, some of the anti-AI people (AAI) commented stating I shouldn’t be using AI. Some were more passionate than others, but there’s clearly a shared feeling of frustration and injustice around AI use. The main focus appears to be on visual artistic endeavours, such as the Studio Ghibli influenced generations. People are upset that an artist’s creation style is being imitated by AI, allowing others to make their own creations in a similar style at the click of a button. I assume this bothers them because they believe once a style is released to the world, it still belongs to the original creator, and any imitation, AI or otherwise, should seek permission or pay the creator royalties of some sort.

The double standard problem

I got thinking about this, and I started wondering if the AAIs would hold other forms of AI generation to the same standard. For example, Joseph Campbell is the writer who defined and popularised the “hero’s journey” style of storytelling. His work helped generations of writers, novelists, filmmakers etc write compelling narratives using the archetypal hero’s journey format. Will we ever see outrage when ChatGPT generates a video script using the hero’s journey without crediting Joseph Campbell? I doubt it. That goes for most writers actually, as practically every creative text output from ChatGPT will be imitating one author or another.

How about when ChatGPT helps someone with a math problem? Should AI credit the 6th century Pythagoreans? I’m aware these examples seem a little silly, but I’m using them to highlight a point. I think it’s valid to consider how far we’re going to go with this. Are we going to stifle progress in the name of tradition? Ironically, I see many of those protesting AI generation appear to be politically progressive, but this call for things to stay as they are, to maintain the old ways for the sake of the old guard, comes across as deeply conservative thinking.

Don’t get me wrong, I don’t see anything wrong with wanting to preserve and develop skill, but in this case it feels like newspaper writers trying to make a moral case against the internet because it makes writing and journalism more accessible to the layman. Or expert filmmakers trying to make a moral case against digital cameras for the same reason. It has tones of the old-school taxi drivers who know their city inside out protesting against Uber drivers who simply use Google Maps and the Uber app to taxi their customers around.

But what’s most concerning for me is how this anti-AI rhetoric could negatively impact animal advocacy.

What this means for advocates

As an animal advocate, advancements in AI have significantly improved my workflow. AI generative technology has not only sped up my process, but given me new innovative ways to create engaging content. Are we to revert to a more cumbersome, time-consuming and ultimately less effective way out of some notion that AI is unfair?

I feel this would be akin to demanding all films are made without the use of digital cameras, or demanding all creative writing and journalism be printed and never online. It feels like an attempt to limit artistic creation and expression only to those who possess the exact skills the already-skilled want them to have, rather than opening it up to a more diverse range of skills and talents.

So rather than just talk about why the outrage feels misplaced, let me show you how AI can actually be a massive asset in the fight for animals.

How can you use AI to help animals?

The best impact to effort ratio you’re going to get as an advocate is on social media. Something that takes you a few minutes has the potential to reach millions of people, and AI can make it even easier to achieve this.

I’m practically a one man team. I create content for Substack, YouTube, Instagram, TikTok, X, Facebook and RedNote, all on my own. I do my own research, write scripts, articles, do my own graphic design for thumbnails, and film and edit my own videos. I also have other projects that I work on in teams on top of that. ChatGPT has been a godsend for me.

ChatGPT has a free plan, and can help with almost everything you’d ever need to create impactful pro-animal social media posts. Here are some image generations you could try on ChatGPT for impactful activism:

• Replace the cow in this photo for a dog (upload an image of another animal in a farm or slaughterhouse for reference).

• Generate an image of someone smiling and wearing a blindfold while eating a burger, but the burger contains a cow’s eyeball.

• Generate an image of a butcher cuts animal diagram but for a dog, use these cuts: Shoulder, Rack, Loin, Leg, Shank, Breast, Neck.

  

  Original image

  

  Replace the cow in this photo for a dog

  

  Generate an image of someone smiling and wearing a blindfold while eating a burger, but there is a cow’s eyeball in the burger

Generate an image of a butcher cuts animal diagram but for a dog, use these cuts: Shoulder, Rack, Loin, Leg, Shank, Breast, Neck.

Sometimes the image generations will require a bit of “massaging” to get them looking good. For example, my butcher cuts for a dog generation didn’t work so well the first time, so I uploaded a real image of a butcher cuts diagram for a lamb and told ChatGPT to swap out the lamb for a dog. This worked well.

Here are a few other examples of how ChatGPT can help you with effective activism:

• Generate 20 pro-vegan posts for X. Focus on exposing factory farming, animal ethics and influencing people to start making changes in their lives to help animals. They need to be highly influential and not divisive. Make them under 30 characters.

• Generate a 10 minute video script for a YouTube video dealing with the common anti-vegan argument “plants have feelings too”.

• Give me a concise, effective response to this person’s comment (paste an anti-vegan comment).

Aside from generating images and text that can help with your creative activism, ChatGPT can also help analyse and improve your strategy. You could ask it to create an activism plan based on what’s most likely to influence people and create change, favouring your interests, skills and preferences. You can run any of your ideas by it, asking it to analyse and tell you whether it thinks they’d work and why. You can even copy and paste your online vegan arguments and debates into it to get a breakdown, analysis, and advice on how you can improve. I actually created a custom GPT called the Convincing Debate Analyser to help with exactly that.

It goes without saying that all AI outputs should be double and triple checked for accuracy, but I hope you can see just how powerful these AI tools are. These six examples alone could save hours of time and money. Most animal advocates are low on both, I’m sure I don’t need to tell you that. Using AI is like having a team of extremely smart and capable people who are always available to help, no matter the time of day, and who can significantly level up the efficiency and effectiveness of your advocacy. AI can help you make a bigger impact, bringing us closer to a better world for animals.

Who knows how this will pan out, maybe there will be strict limitations applied to AI generators, maybe not. The debate is ongoing. What I do know is that if you’re an animal advocate, I think you should be using all the tools that can help you make a difference for animals. The fight for animals is already hard enough. If technology gives us an edge, if it helps us reach more people faster, we should use it. What do you think?

Note: Since writing and publishing this post, a few have raised concerns about theft, and the environmental impact of AI. I’m adding this short segment to discuss these points.

Isn’t using AI generations theft?

AI-generated text and images are always based on what the model has seen and processed. This is the nature of AI generation. It can make combinations of things it has seen to create entirely new images, but they will always be consisted of things that exist or have existed. Any movie script you generate will have been trained on existing stories, writing styles, structures etc. Any image you generate is built from existing photographs, illustrations, or drawing styles. Some are calling this theft but, interestingly, that’s exactly how it works for humans too.

Like AI, the human brain can also only generate things it has already seen and processed. Any face you see in a dream is likely a face you saw in real life, or an amalgamation of faces you saw in real life to create a new face. Anything you create will have been influenced by what you have seen, whether that be another artist or writer, or just something you saw or heard while walking in the city.

Taking someone’s work beat for beat and replicating it exactly would count as theft in the eyes of many, but taking inspiration from their style and making your own creation in the same style? This has never been considered theft, and for good reason. This is how creativity works—we are inspired by ideas and incorporate them into our creations. Countless filmmakers have “stolen” colour palettes, lens effects, camera angles, dialogue styles and story beats from other filmmakers. Countless musicians have “stolen” time signatures, drum techniques, guitar techniques, amplification modifications and writing styles from other musicians.

Now AI is allowing people to take writing styles they like, as well as photography and illustration styles they like, and easily incorporate them into their own creations. For example, this could be using a Tarantino filmmaking style for an independent vegan film, or using a Studio Ghibli illustration style to bring new life to an old photograph where you’re having a cute interaction with an animal. I don’t believe this is theft, I believe it’s inspired creativity that is now easily accessible for the masses when before it was only accessible after years or even a lifetime of training.

Is AI bad for the environment?

The environmental impact of AI, especially training AI, is high. It uses a sizeable amount of energy to train new models, though this is something that takes place infrequently. But it appears that for general use, for image and text generation, AI could actually be having a lower environmental impact than manual creation.

You can see the carbon footprint estimates for these tasks below, taken from this study.

This figure compares the CO2e emissions of AI and humans engaged in the task of writing one page of text. AI writing (via BLOOM or ChatGPT) produces 130–1500 times less CO2e per page than a human author. AI also produces substantially less CO2e than the computer usage to support humans doing that writing.

This figure compares the CO2e emissions of AI and humans engaged in the task of creating one image. AI image creation (via DALL-E2 or Midjourney) produces 310–2900 times less CO2e per image than human creators. AI produces many times less CO2e than computer usage to support humans making images.

It’s commendable that many have concern for the environment when it comes to AI, but something strikes me as quite peculiar. Many are demanding that people stop using AI to save the planet, but they aren’t demanding that people stop leisurely using social media, streaming services and the internet in general to save the planet.

Not only that, but they themselves have most likely been leisurely using social media, streaming services and the internet in general for years, if not decades. This general, leisurely use of the internet is unnecessary and has an incredibly high environmental cost—if we’re going to demand people refrain from using AI for the planet, surely we should be consistent?

Of course, consistency here would mean a huge loss of support. People would reject these ideas en masse, which is probably why no one is demanding it. Understandable, but it does suggest that the anti AI movement is more about reverence for Japanese artists and fear of technology, as opposed to genuine concern for justice and the planet. For example, some critics of the anti AI movement have highlighted that there wasn’t much, if any outrage when people were “Disneyfying” themselves, or using the Pixar filter.

I encourage all animal advocates to start exploring this new AI technology immediately. The possibilities are endless, and the barrier to entry is the lowest it has ever been. I believe we need as many advocates as possible using these tools to innovate, reach millions, and help millions of animals.