The Nuance of AI: Understanding Its Impact and Complexity

Michael R Kiel
Apr 4
5 min read

Updated: Jul 7

There’s something heavy in the air when we talk about AI...

For some, it brings hope—new tools, access, and ways to create. For others, it evokes grief, fear, even anger. For many of us, it stirs a confusing mix of emotions.

This post isn’t here to defend or condemn AI. Instead, I want to offer something that often gets lost in fast-moving, emotionally charged conversations, especially while we’re online: space.

Space for Reflection

I invite you to stay curious—maybe even open-minded—as we walk through each of these discussions. Take your time; avoid rushing to conclusions. There’s no need to skim for a gotcha moment. Let’s sit with the complexities together, holding them with care and pausing for critical reflection.

1. AI and the Environment: A Bigger Picture

Let’s start our mindful examination of AI with something tangible: carbon, energy, and water.

Training massive AI models, such as ChatGPT-3, comes with significant costs. A widely referenced paper (Energy and Policy Considerations for Deep Learning in NLP, Strubell et al., 2019) estimates that training a model like GPT-3 can emit over 550 metric tons of CO₂—roughly equivalent to the emissions of 120 gas-powered cars over a year. 👉 View the study on ACL Anthology.

Moreover, this figure does not account for water consumption. Cooling data centers for these models uses vast amounts of clean water—up to 700,000 liters during training cycles, as per industry reports.

These numbers matter, but so does context.

Here are five frequently used services and their (often overlooked) environmental impacts:

Streaming video: About 30 minutes of streaming can emit 1.5–2.4 kg of CO₂, primarily driven by data center demands (Guardian, Oct 2024 , CarbonTrust.com).
Personal vehicles: The average gas-powered car in the U.S. emits around 4.6 metric tons of CO₂ per year—more than the total emissions of many AI systems throughout their entire lifecycle (EPA, 2023).
Household water usage: The average American household uses approximately 300 gallons of water daily, amounting to over 100,000 gallons per year, mainly for landscaping, laundry, and showers (USGS, 2023).
Bitcoin mining: Globally, Bitcoin consumes more than 100 terawatt-hours of electricity annually, which is more than entire nations like Argentina (Cambridge Bitcoin Electricity Index).
Fast fashion: The clothing sector contributes to up to 10% of global carbon emissions and causes significant water pollution and textile waste (European Parliament).

These figures do not excuse AI’s environmental footprint, but they do prompt us to adopt a more critical and balanced perspective. If we’re going to call out harmful systems, let’s scrutinize all of them equally.

We often express outrage where we feel the least empowered. AI is sizable, enigmatic, and rapid—making it a perfect target for that frustration. However, it’s not the only entity that deserves thoughtful accountability.

2. Is AI Theft? The Ethics of Training and Style

A deep tension exists here, touching on consent, ownership, and the essence of creativity itself.

Most large generative AI models are trained on enormous datasets taken from the web—artwork, articles, images, and books, often scraped without permission. Many creators never granted their consent, and some didn’t even realize they had the option to opt out—because they couldn’t.

This has led to a rise in lawsuits, such as:

Authors like Sarah Silverman filing claims against OpenAI and Meta regarding copyright infringement.
The New York Times suing OpenAI, asserting its journalism was used without a licensing agreement.
Artists experiencing AI-generated works mimicking their styles or even displaying their names in counterfeit signatures.

This isn't a conspiracy; it’s a genuine issue taking place right now.

There’s also a scientific concern. A 2022 paper (Quantifying Memorization Across Neural Language Models, Carlini et al.) revealed that large language models can memorize and reproduce exact phrases from their training sets. 👉 Read the study on arXiv.

The legal landscape is still evolving. Historically, we have faced similar challenges.

Photography, music sampling, and digital collage all sparked controversy upon their introductions. Over time, copyright frameworks have adapted to accommodate both protection and innovation, for instance:

The Anxiety of Photography, published by Aspen Art Press and the University of California Press, discusses how early photography was viewed as mechanical and threatening to painting and illustration. Over time, it gained both legal protection and artistic acknowledgment. 👉 The Anxiety of Photography (JSTOR).
Regulating Digital Sampling: A Comparative Analysis of U.S. and U.K. Copyright Law, published in the Journal of Intellectual Property Law & Practice (Oxford University Press) examines historical legal disputes surrounding sampling, including landmark cases like Grand Upright Music v. Warner. This shows how courts began moving toward more nuanced approaches, including licensing standards and limited fair use. 👉 Oxford Academic - Regulating Digital Sampling.
Appropriation Art and Fair Use: Campbell v. Acuff-Rose and the Future of Parody, Satire, and Pastiches in the Columbia Journal of Law & the Arts discusses how digital and mixed-media art, particularly collage, faced copyright pushback. This analysis highlights how U.S. courts have employed the transformative use doctrine to shield some forms of digital collage and remix art. 👉 Columbia Law - Fair Use and Appropriation Art (PDF).

These instances illustrate that creative disruption is not a novel phenomenon. Time and deliberate legal evolution have shown that it is indeed possible to safeguard original work while making room for new forms of expression. As we navigate AI’s influence on creativity, we build upon lessons we've already learned.

However, the ethical line remains clear: Consent and credit matter. They always have.

3. AI and Illustrators: Grief, Change, and Possibility

This aspect feels deeply personal.

Illustrators are witnessing the disappearance of clients, dwindling contracts, and portfolios becoming mere references for models that can replicate their life’s work in seconds.

This situation is real. It breeds pain and deserves acknowledgment.

According to Adobe’s Creative Professionals and Generative AI study, 32% of artists fear losing their jobs to AI. Concurrently, 71% foresee using generative tools to enhance their work. 👉 View the Adobe study summary.

The World Economic Forum’s Future of Jobs Report 2023 presents a similar dichotomy: 83 million jobs may be displaced due to automation and AI—but simultaneously, 69 million new ones could emerge, many in creative-adjacent or hybrid roles. 👉 Read the WEF report.

The reality is this: this is disruption, not disappearance. However, disruption can be painful, especially when it arrives uninvited.

That’s why I’m choosing a middle path:

I will soon be collaborating with real illustrators on specific projects—affording them the opportunity to create a cohesive visual narrative over the course of 2–3 months.
I will feature genuine nature photography on Wednesdays and Sundays, always ensuring credit and transparency.

These may seem like modest gestures, but they convey a crucial message: You matter. Your eye matters. Your voice matters.

4. Holding Complexity Is a Skill

In Acceptance and Commitment Therapy (ACT), we often say:

“You are not your thoughts.”

Moreover, I would assert:

"You are not your fear of the future. You are not your discomfort with change. You are not your outrage."

"You’re something deeper."

"You are your values."

This means you can:

Embrace new tools while still asking tough questions.
Care for the environment while learning about and using AI judiciously.
Support artists while still creating in a hybrid world.

We don’t need to reduce every issue to “for” or “against.” We can endure the tension. Mastering this is a valuable skill, and one we should strive to develop.

Closing Thoughts

I don’t have a viral call to action for you here. No thread. No checklist. No oversimplified stance.

I simply hold this belief:

Nuance still matters. Slowing down is powerful. And curiosity, when grounded in values, is more sustainable than outrage.

Thank you for making it to the end of this post. I appreciate your time, openness, and willingness to grapple with discomfort rather than shout over it.

This is our collective effort, and we are in it together.

Michael R Kiel, MA, LPC