AI is replacing artists and destroying the environment? – A deep dive into AI ethics and misconsceptions

I have to be honest, until my recent participation in the TikTok Halloween AI trend (which has “sparked some debates” to say the least), I did not consider this topic as something so controversial as it turned out to be. But since I come from a pretty tech savvy family background, I have seen people around me work with and adapt to new technology all my life, which made me stay pretty open-minded to new tech in general and also has taught me not to face them with fear. I understand now that for some people, this can feel very different – and I really get where those concerns are coming from.

And while I am clearly no expert in AI or tech in general, I always do my own research when controversies arise. I’m not a fan of one-sided, polarising content and black-and-white-thinking. Throwing around oversimplified slogans I picked up on the internet and bashing something I don’t fully understand, won’t make a positive change in the world at all, if anything it’ll just spread hate and further misinformation.

So, what did I do instead? – I took the claims and fears I kept reading online regarding AI seriously and took the time to do a deep dive into AI ethics and research to what extent those claims are actually true, and, most importantly, how me and anyone else wanting to use AI, can do so in a more responsible way!

Now I will have to say this: A lot of the negative things I have been reading regarding the use of AI, are based on real concerns that got oversimplified and turned into slogans. So I do get the overall concern. And even though I have gotten less critical, the more educated I became on the topic, I myself am still critical about some aspects of AI use.

However, while those claims have a grain of truth, they’re usually thrown around without nuance. Some are either just objectively false or aren’t made from a neutral or fully educated perspective and (which is common in any new technology) people are scared of something they don’t fully understand and that they now feel threatened by.

But rejecting AI entirely might mean missing the chance to shape it in a better direction. Just like any tech, it’s about how it’s built and used. This is why I will take the chance to not only educate myself, but also others who might come across this controversy, and it is exactly what led me to create this article.

To be clear: This is by no means supposed to be simply “AI defense”, or any form of justification for the unethical aspects some AI models or their use might have.

It is simply meant to be educational, eliminate prejudices and to ease fears. To give you the chance to see how many different aspects need to be considered in the discussion about ethics in art and creation in general and how there is not a simple answer of wrong or right but art has always been bending rules and existing in grey areas.

In fact, a lot of art forms are technically illegal or ethically messy but are tolerated or even romanticized, because they don’t directly threaten the audience’s sense of identity or income. But when artists see AI touching their world, suddenly the moral line gets redrawn.

That doesn’t make the anger invalid – it’s just human nature. We tend to accept the grey as long as it benefits us.

This “ethics-by-emotion” phenomenon is why discussions around AI art often sound moralistic rather than philosophical, and it’s exactly what real AI ethicists try to move beyond.

In fact, this concept is actually pretty similar to my own morals I like to live by. I like to view discussions from the most neutral perspective possible, taking a step back to understand both sides and checking the facts before forming an opinion, rather than reacting emotionally. Following discussions online, I often see others pick and choose by what benefits them in that moment, but don’t live by the same morals in other aspects of their life. I personally try to avoid that as much as possible and try to be honest with myself when I fall short. I also try to accept that I can’t always live up to every ideal, instead of constantly chasing an impossible standard. On top of that, worrying about and fighting things I can not possibly change, is something I will always try to avoid. And let’s face it – AI isn’t just gonna magically disappear by bashing companies or individuals who are using it or by covering our eyes and pretending it doesn’t exist.

Anyway, let’s get into this topic and look into some common claims regarding AI and the truth behind them!

“AI is stealing from artists.”

So, this is only true to some extent and also depends a lot on your own ethics regarding copyright. Some early image generators (like Stable Diffusion or Midjourney) were trained by using huge datasets of online images – often scraped from sites like ArtStation, DeviantArt, Pinterest, etc. Many of those images belonged to artists who never gave consent for their work to be used in training. So from their perspective, it feels like their style and labor were used to create a competing tool – without credit, permission, or payment. AI models however don’t literally “copy” or “steal” artworks. They don’t store or reproduce specific images pixel-by-pixel; they learn abstract patterns from data, then generate new combinations. It’s not theft in the legal sense – it’s more like a remix trained on collective visual data.

Just like a human artist will learn by looking at art from other artists and recreating parts of those artworks to get better until they finally develop their own style, AI does something very similar. In the end, both humans and AI, could recreate art (even beyond images) in the style of another artist or featuring original creations by other artists, and not even credit them, claiming it as their own. AI might just be a little faster at doing so. Some might even profit off of it. – Which, yes, is absolutely unethical, but it has nothing to do with AI as a technology and everything with the way people use it.

And, to be clear, I as person do not support this, regardless of who made it, humans on their own or by using AI.

Also if we’re thinking about the whole issue of copyright, it is important to mention that art has always lived in grey areas.

Look at these AI generated pictures of me next to popular horror movie characters. If I photoshopped these by myself, for fun, for my own private use, without making a single dollar from it, would you consider it theft or a fun fan creation? Would you be okay with me posting or even selling it, if you created/ owned the rights to this character? Why does it matter which tools I used to create this?

Is fanart or any edit made by hand that includes characters or other things from popular media, or anything you technically don’t have the rights to, ethical to you?

Now think twice – would you bash a photographer by saying that traditional art is the only “true form” of art and they shouldn’t use a camera as a shortcut, but instead learn how to paint or hire someone to do it for them? Would you tell them they’re ruining the lifes of traditional portrait painters? Or has photography as an art form evolved on its own during the years and just became a widely accepted part of the art community, coexisting with other, previously existing art forms? What about digital art?

Is art only about the skill of creating in a specific way, or the power of imagination itself, no matter how you bring it to life?

In my opinion, this debate is not much different to the discussion about AI nowadays.

And what about other aspects of ownership?

Let’s stick to the example of a photographer. If I go on an urban exploration journey to take some epic photographs, I would very likely be trespassing, as I would enter someone else’s property without their consent. Doesn’t this also make my art unethical?

You could also ask yourself the same questions about other creations such as fanart as mentioned before, cosplay, DIYing items from stores you can’t afford to buy from, music sampling and remixing, scanning and 3D printing vs modelling, collage/ appropriation art (think Andy Warhol, Barbara Kruger, Richard Prince who all used preexisting works like ads, photos, logos), street art/ graffiti, the list goes on.

As you see, while it is absolutely important to have this conversation about ethics in art and creation in general and to improve in every aspect of this, art being controversial or some sort of grey scale is nothing new.

We’ve always tolerated art that bends rules. The outrage only spikes when people feel the impact. But ethics should be consistent, not reactive.

Each and everyone of us has a list of things others might consider unethical, none of us are perfect. Claiming to be the exception doesn’t make us moral, it would be unrealistic – even arrogant – and it just blinds us to our own flaws. Our goal and way to a more ethical life, should be finding middle ground, not striving for perfection.

Also in the end where do we draw the line? Inspiration and some sort of “copying” from other sources have always been a huge aspect of art in general. And as long as it isn’t an obvious 1:1 copy of someone else’s creations, to the point where you are barely able to tell apart the original from fake, this has always been accepted as a norm in the art community. (Not saying debates about other grey areas don’t exist.)

Now, speaking about this, ethical AI models with consent-based training exist and we should aim for improvement of those models and proper regulations of AI training in general, instead of demonizing the technology as a whole. Models trained only on licensed or public domain data, or with artist opt-in systems, don’t exploit anyone and should become the standard in future AI training (e.g. Adobe Firefly, Getty AI, Fair Diffusion).

But we will get into this topic a little later when we talk about what we can do to make AI use more ethical!

“AI will replace all human jobs.”

While AI will automate tasks, it rarely replaces entire roles outright. It often changes how jobs are done and can create new kinds of work too (such as AI ethics, prompt design, training, auditing). As technology evolves, it has always taken over simpler tasks, allowing humans to concentrate on what truly requires creativity and judgment – AI simply continues that pattern.

“AI can think or create original ideas on its own and therefore will replace real artists.”

In general, every new invention feels disruptive at first. Art has survived photography, digital design, and Photoshop – this is just another wave.

AI isn’t capable of independent thought or original ideas, it can only work with the material and guidance humans give it.

An interesting study by the Cornell University shows, that blending artistic expertise with AI tools can enhance creativity and support artists as they navigate the digital age. While AI can fill in gaps and sometimes might even outperform humans, it can not do so without their input and guidance, making it a powerful tool in supporting and amplifying human creativity and artistic vision.

Just look at the development from traditional to digital art. I remember how the raise of digital art caused a very similar debate. Today both traditional and digital art coexist, still most people will give more value to traditional art as it is often more time consuming and more expensive to keep buying materials.

AI isn’t the first technology to disrupt creative work. Everything it “knows” and “uses” comes from human creation, so its potential is shaped entirely by us. It cannot generate emotion or deeper meaning on its own – it can only help bring an artist’s vision to life.

It is also already used in many different aspects of life and it certainly is nothing new!

A few examples being:

Autocorrect, movie animation and special effects, photoshop tools, filters and templates in social media and editing apps, old school “artsy” filters, beauty filters, auto generated subtitles, tools to help disabled people, etc.

While certain sectors may experience some revenue losses due to AI-generated content, so far (as of 2025), there are no large-scale studies or labor statistics showing a measurable decline in creative employment caused specifically by AI.

Artists have been struggling long before AI’s boom because of broader economic issues:

• Pandemic (2020–2022): Massive loss of creative jobs – galleries closed, events canceled, film and live art halted
• Inflation & cost of living crisis (2022–2025): Less disposable income = fewer commissions, lower budgets.
• Wars, political instability, and recession fears: Businesses cut marketing, entertainment, and art budgets first
• Arts funding cuts: Many countries slashed culture budgets post-pandemic
• Platform algorithm shifts: Organic reach on Instagram/TikTok dropped; artists struggle for visibility regardless of AI

All of these factors have far more quantifiable economic impact than AI does. AI is also creating new roles and fields within the creative industries, offering opportunities for artists to adapt and innovate.

Artists aren’t losing jobs because “AI replaced them.” They’re facing the ripple effects of a global economy hit by a pandemic, wars, inflation, higher costs, shrinking budgets, and cultural underfunding. Art always thrives when things get better – and struggles when people can’t afford to value it. Artists are affected by the same economic shifts hitting every industry. When budgets get tighter, creative work is often the first to be cut.

Embracing AI tools can enhance an artist’s creative process, while still maintaining a balance between technology and artistic authenticity. Artists, like any other professionals, need to continually update their skills and stay informed about technological advancements. Adapting to new technologies can help them find new opportunities and remain competitive. In the end, technology isn’t going away. Complaining about its existence won’t undo it – the only way forward is to learn, adapt, and find ways to use it to your advantage. Fear doesn’t protect your art, but creativity and adaptability do.

Also, a lot of creative jobs have evolved since the very beginning of art due to new technology. Each and everyone of those more traditional artists had the same fear of being replaced by those technologies, but in the end, the way those jobs were done simply changed and it will be just the same with AI.

• Portrait painters → Photography
• Handwritten books → Printing presses Scribes became editors, typographers, illustrators. Calligraphy returned later as an art, not a necessity.
• Stage Actors → Film Actors → Motion Capture Performers Each medium changed the craft: theater to cinema to CGI.
• Orchestral Musicians → Electronic Producers / DJs Synthesizers, drum machines, DAWs, reshaped music. Many traditional musicians feared “machines” would ruin art – instead, whole new genres (EDM, ambient, lo-fi) were born. Today, you see hybrids – live instruments with electronic layers.
• Darkroom Photographers → Digital Photographers → Mobile Creators Every step made the medium more accessible. Professionals adapted by specializing – editorial, fine art, retouching, drone photography, etc.
• Seamstresses → Fashion Designers → 3D Fashion Artists Sewing by hand shifted to industrial machines, then to CAD and 3D modeling. The job evolved from making garments to conceptualizing collections and digital fashion design.
• Newspaper Layout Artists → Graphic Designers → UX/UI Designers The core skill – visual communication – stayed the same, but tools and contexts changed.
• Cartographers → GIS Specialists / Data Visualization Designers Hand-drawn maps → satellite imagery → interactive data maps. The artistry shifted into information design and storytelling with data.
• Traditional Illustrators → Digital Artists → Concept Artists / AI-Assisted Artists Wacom tablets, Procreate, Blender, and AI tools changed workflows. The creative direction and vision still come from humans – tech just expands the toolbox.

Every new wave of tech lowers barriers to entry, shifts the focus of creativity and forces artists to adapt, not vanish. It also offers countless possibilities to make art and various other aspects of life more accessible to everyone!

In the end, photography didn’t kill painting. Synths didn’t kill music. AI won’t kill art – it just shifts how we create.

“AI destroys the environment”

The truth behind it:

Training very large AI models uses a lot of computing power – which means high energy consumption and carbon emissions, especially if powered by fossil fuels.

For example, training GPT-3 reportedly emitted as much CO₂ as a few cars over their lifetimes.

The misconception:

People sometimes act like every single AI prompt burns a rainforest, which isn’t true.

After training, running a model (e.g., generating art, text, chat) is relatively energy-efficient, often less than streaming a few seconds of video or running a gaming GPU.

Plus, newer models and data centers are becoming greener, using renewable energy and optimization techniques.

So while it’s true that training big AI models uses a lot of energy, so do many digital tools we already rely on, like streaming, watching shows, gaming, video calls, crypto, social media, and even storing your photos. – Would you also give up all of those?

To get back to the car example, I did the math for you – Instagram emits roughly the same CO₂ in one day, as about 380 average cars over their entire lifetimes. So next time, maybe think twice when you talk about AI destroying the environment on a platform that also contributes massively to the problem.

Taking it a step further, what about other, way bigger contributors to CO₂ emission? In just one week, global meat production emits around 136 million tonnes of CO₂-equivalent, roughly the same as 2.5 million cars over their entire lifetimes. – Well, ever thought about going vegan?

Reality is, the real solution isn’t avoiding tech – it’s advocating for greener infrastructure, renewable-powered servers and energy-efficient design in any industry out there.

What can we do to take a step towards ethical AI and how can we contribute?

• Educate, don’t hate – advocate for consent & credit
• Share info about HaveIBeenTrained.com, where artists can opt out of certain datasets
• Support ethically trained image models, such as Fair Diffusion (by Spawning.ai), Getty Images + NVIDIA (2023), Adobe Firefly, Leonardo.Ai (Premium Models like Alchemy)
• Ethically trained text models, such as Anthropic’s Claude, Mistral (open models)
• And ethically trained audio / voice models like ElevenLabs (Voice Library & Opt-In Training)
• Choose platforms that are transparent about their data sources (e.g. Adobe Firefly, Getty AI, Fair Diffusion)
• Look for models trained on licensed, public domain, or opt-in data
• Support open letters, petitions, or orgs pushing for AI transparency laws (like the EU AI Act or creative rights coalitions)
• If you use AI art in your work, credit the tool and your role – be clear it’s co-created, not “painted by hand”
• Support human artists, keep commissioning artists, buying prints, and sharing work you love, give shoutouts and credit to the styles or creators that inspire you
• If you use AI, blend it with human creativity – treat it like a collaborator or reference, not a replacement
• Support research into energy-efficient AI (like smaller, open models or edge AI)
• Be transparent about your own use
• Remember: “Ethical AI” = Culture Shift, tech mirrors values of its users. We set them.
• Give the bigger companies time to improve and take the right steps

There’s a lot of hype and fear around AI. A lot of uneducated black and white thinking. But it’s really just a tool. Like any technology, it can be used responsibly or recklessly. The key is how we design and regulate it.

Companies around popular AI-Tools are already taking steps towards the right direction. For example, OpenAI (ChatGPT/DALL-E) has removed artist-style prompts (e.g. “in the style of [living artist]”) to reduce style mimicry and potential copyright violations. They also have strong content filters and ethical guidelines – no deepfakes, NSFW, or harmful content.

Google Gemini (formerly Bard + Imagen) is also avoiding “style mimicry” by banning prompts using artists names. They’ve also made real commitments to responsible AI, with dedicated ethics teams and transparency reports.

And while companies are improving, I still hope to see even more progress towards transparency and consent-based AI training!

In my personal opinion, using AI in general is totally fine, it’s a fun and powerful tool with room for improvement and occassionally I like to play around with it, just like I enjoy other forms of “grey area art” (even if I maybe shouldn’t). As long as you use it for yourself, for fun or inspiration and as long as you are open about what parts of your work have been created by or with the help of AI tools, I think it is really not that deep. I also will continue to create myself and to support small artists, creators and businesses the same way I did before the rise of AI.

I draw the line once you make money or significantly profit off creations, that are visibly based on the work of other artists (regardless of it being through unethical use of AI, or by someone copying another person).

Profiting off the creations of someone else in any way (and/or claiming them as your own) is not okay and I try to hold myself to this standard. This is why I would for example never use random AI art for sponsored or paid posts and also would never pay for fanart, while I think it is still fine to create for my personal use. I will also always credit other artists if they inspired a makeup look or another form of my art.

In summary, supporting human art and ethical AI can (and will have to) coexist – it’s not either/or.

If you want to keep reading about this topic, I highly suggest to check out this article by the University of Oxford about a study they did regarding the replacement of artists by technology.