Why AI: Finding a middle ground in the AI debate

This is the first in a series on AI and my journey using it as a multimedia storyteller and journalist.

I remember when, between 2023 and 2024, early users of a new breed of generative artificial intelligence (AI) and large language models couldn’t stop talking about it. A group of fintech guys I knew raved about AI chatbots at a dinner. Co-workers recounted their AI programming experiments during meetings. My social media feeds were peppered with fellow media professionals lowkey flexing the images they created with their exclusive, early access to tools like DALL-E and ChatGPT. During that period, AI was all the rage among the tech and media elite. But I wasn’t interested enough to figure out how to get the invites, find a way around intrusive verification requirements, or fully understand what it could do for me. So I put it off.

Then ChatGPT and a few other chatbot tools became more accessible to the public, with great fanfare, leading to an explosion in use. I dabbled but didn’t really get involved until a huge career change in 2025 allowed me to explore frozen passions I’d had for years. First, I set out to finish building Klatchmaker, a friendship-making website I’d always hoped to launch, and began experimenting with AI to debug the old code. A few months into this mission, I found myself thinking about the film career I once pursued but forfeited for journalism years ago. That’s when the idea dawned on me: AI could be used to make movies. Even though it wasn’t that long ago, the idea was novel at the time. The campy video of Will Smith eating spaghetti was still just a silly meme. One of the most recognized examples of a “realistic” AI short film was an uncanny video about a woman with dementia who finds herself in a time-traveling diner (where she also eats spaghetti). I saw this as an opportunity to reconnect with a passion, eventually becoming just as excited about AI as the tech and media bros had been a year or two earlier.

But of course, true to form, by the time I got around to fully embracing AI, the public excitement took a turn. Ominous media coverage zeroed in on the dangers of AI. News outlets demonstrated how easy it was to create deepfakes. Tech publications that once raced to cover the latest and greatest AI model releases, emerging AI companies and early-access invites now shifted coverage to focusing on the risks AI posed to society.

Meanwhile, Hollywood unions made AI replacing or duplicating artists a primary concern during their negotiations with studios, winning over much of the public in the process. Media organizations sued AI companies for using their articles to train AI models, alleging copyright infringement. School cheating scandals, bullying and incidents involving self-harm became high-profile AI cautionary tales. Headlines quoting tech CEOs bragging about replacing jobs with AI enraged people, while talking heads and podcasters crafted hot hot takes that armed the public with anti-AI talking points on AI’s impact on jobs, the environment and copyright.

Amid all of this, a portion of the once-enthusiastic tone surrounding AI, particularly generative AI (using AI to create media), is becoming increasingly vitriolic. The press started calling lazy generative AI cash grabs “AI Slop,” which anti-AI internet users have since co-opted and now use as a dismissive pejorative for all things generative AI. I see the hostility on a regular basis: I moderate an AI filmmaking subreddit and increasingly have to remove comments using the terms “slop” or “you people,” along with other choice words clearly designed to punish people sharing AI-assisted work.

While anti-AI critics make several valid points about AI’s ills (some of which I address below), it’s becoming clear that a blanket rejection of AI and efforts to harass and intimidate people who use it might actually cause more harm than those joining the mob realize. AI can have positive uses. Since embracing the technology, I’ve come to realize that instead of demonizing AI and all the people who use it, it might behoove us to focus on how people should use AI in a responsible way.

In this piece, I explore the main reasons why.

AI is not just here to stay; it’s there to stay.

Some of the American backlash toward AI can be narrow-sighted. Among countries with the highest GDP per capita, the United States has one of the lowest AI adoption rates, 31 percent, according to Microsoft’s Global AI Diffusion report. Singapore, Ireland, Norway, Switzerland and Qatar have much higher rates, up to 70 percent.

AI is not an American or Western phenomenon. Much of the world is using it, and some countries are embracing it more than others, giving themselves a leg up in AI innovation and AI’s economic opportunities. Several Asian countries are among the fastest-growing AI economies, the Microsoft report showed. According to the 2026 AI Index Report from Stanford University, South Korea, Luxembourg and China issued more AI patents per 100,000 inhabitants in 2025 than the U.S. Meanwhile, China, a leader in AI research, is often cited as the U.S.’s biggest competitor in AI dominance.

By pressuring Americans to shun AI, critics could make the U.S. less competitive in an industry that is young enough to invite all sorts of competition, innovation, and advancements. Discouraged Americans risk being left behind as the industry becomes more established globally.

That brings me to my next point.

If you don’t learn AI, you risk being passed over, outperformed

Declining to participate in the AI space can make people less competitive in professional spaces, which can be be detrimental to some of their livelihoods. If, for instance, a manager wants to hire a freelance developer, they will likely ask a few potential freelancers for rates and quotes. Let’s say Freelancer A refuses to use AI and quotes a fair rate based on the amount of time it would take them to code the project by hand. Meanwhile, Freelancer B offers the client the same rate but can accomplish the task in half the time as Freelancer A because they are using AI to scaffold and debug their code. How many rejections will Freelancer A get before they realize the problem? If given the choice, many companies are going to choose the freelancer who can accomplish the task more efficiently, whether they realize that person is using AI or not and whether they realize the person they passed over may have refused to use AI.

A similar scenario can apply to co-workers, in which employees who use AI outperform teammates or deliver projects more efficiently, thus raising expectations (of course, this is assuming its use is within the employer’s guidelines).

Discouraging or intimidating people from using AI in a world in which AI is already giving peers a competitive edge can therefore become professionally constraining.

Unfair barriers for women, people of color

Anti-AI rhetoric is especially threatening to women and people of color interested in AI. Actress Reese Witherspoon has become a vocal advocate for women and girls using AI because, as she puts it, “We [women] don’t want to be left behind.” Witherspoon has gotten some backlash for her beliefs, but the data backs her up. Even though women represent half of the U.S. population, they account for only 36 percent of U.S. STEM field workers, according to the National Science Board. Discouraging women from participating in AI could reinforce this trend. Witherspoon especially discusses the rise of AI in Hollywood and not wanting women to be left out of it. As someone who moderates an active subreddit for AI filmmaking, I can attest that this space in particular is already decidedly dominated by a relentless male gaze.

I would also expand Witherspoon’s argument to include people of color, of course. Statistically, Black and Latino Americans are also underrepresented in STEM fields, accounting for 9 percent and 16 percent of the American STEM workforce, respectively. AI, still in its early, developmental and experimental phases, is an open opportunity for BIPOC (Black, Indigenous, and people of color) and women to step through the door early, with everyone else, and not get left behind again.

While STEM diversity gaps can be linked to systemic and structural sexism and racism, the newness of AI lowers, but doesn’t eliminate, many of those barriers. As AI becomes a multi-billion dollar industry worldwide, women and BIPOC have a chance to no longer be left behind. Critics who vehemently pressure people to avoid taking part in an emerging tech field might not be considering this data.

Hitting the wrong target

Anti-AI vitriol often implicates the wrong people. In addition to pressuring women and BIPOC out of the space, AI backlash makes it harder for people struggling economically to get ahead.

This is a curious issue: AI, of course, has been around for decades, used in film, research, tech and other fields. When IBM’s Watson defeated trivia champions on Jeopardy! in 2011, the stunt was called a wonder. Meanwhile, AI-based software used to create digital crowds in the original Lord of the Rings films has been viewed as an industry game-changer. Even though Watson and earlier forms of artificial intelligence had their critics, it was not until companies made the power of generative AI available to everyday people to do things they wanted that everything AI-generated became labeled “slop.” Specifically, it was not until its availability transcended exclusive tech and media beta testers that mass panic and fury ensued.

Too often in the AI debate, regular, everyday, low-income and/or marginalized people become proxies for out-of-reach corporations. They are low-hanging fruit, often the easiest targets of the public’s AI backlash. Yet a lot of the arguments against AI center around corporate issues that everyday people neither control nor create. Take AI filmmaking, for example: Many of the creators experimenting with AI filmmaking are regular, independent, non-industry people who oversee no filmmaking jobs to replace. If they were not creating AI films, there would be no actor hired or any job saved. Yet, ironically, many of these creators become cannon fodder for anti-AI critics, even while Hollywood-backed AI projects receive media attention and appear at festivals and such. Netflix reportedly has more than 300 titles using generative artificial intelligence on its platform, and people, possibly including those who will refuse to give an independent AI movie the time of day, still watch them.

Hollywood studios and leaders (which have already been using AI for years) can absorb some AI-related criticism. But everyday, regular people creating AI-assisted work, many of whom have been structurally and systematically shut out of employment in Hollywood and elsewhere, are the ones often being told not to use AI. It’s not easy to get a movie made in Hollywood, especially one told from marginalized voices. In this sense, AI allows people to create storytelling opportunities that either did not exist for them or would not otherwise be available.

In fact, many disadvantaged people may be using AI to improve their lives and circumstances (through education, job prep, entrepreneurship and creative development). Personally, AI was an opportunity for me to explore paths and passions after being left out of the corporate space.

Finding a sensible approach to AI

Making AI taboo will not make it go away. Ignoring it may not be the best approach either.

Assuming AI is here to stay, is being embraced by major economies around the world and may be an opportunity for people to close gender and racial gaps in tech, building an indiscriminate shame culture around AI, rather than focusing on responsible use, could have a negative impact on some of society’s most disadvantaged groups in the long run, including women, people of color and the unemployed and underemployed. At the same time, there are many valid criticisms of AI.

Here is a middle ground: Instead of demonizing AI as a whole, maybe we should discuss how people can use AI ethically and transparently. Could there be a way for women, people of color, visionaries and the unemployed and underemployed to incorporate AI into their lives while minimizing harm to others? This may not be a question that can be answered in one go, but it could be a more productive direction for AI discourse.

Ethics

In the AI filmmaking subreddit, all users must operate under an AI Code of Ethics (inspired by the Society of Professional Journalists’ Code of Ethics). This is an example of embracing AI while still addressing some of its ethical issues. More AI users should operate under similar guidelines. Criticism and accountability should be based on whether people adhere to these principles.

Corporate accountability

Anti-AI criticism that involves corporate issues should be aimed in the right direction. While there are several consumer-related AI concerns, everyday consumers should not be cannon fodder for lazy criticism that should be directed at people with more power and responsibility. Regular AI users should not be shamed for issues that corporate giants refuse to address, nor should they be used as proxies for hard-to-reach corporate decision-makers.

Consumer responsibility

AI users also have a responsibility to be open and transparent about their use of AI, allowing people to assess their work accordingly. Those who do adhere to ethical and transparent principles should not be lumped in with the bad actors.

Education

At least when it comes to generative AI, there are a lot of misconceptions about what it is and how it works. This makes it easy for people to instantly dismiss or demonize it. A healthier approach would be getting people on the same page about how various forms of AI actually work and how and why some are using it. Then, we should use this knowledge to make distinctions between various types of AI users instead of painting them all with the same brush.

Addressing common controversies

But what about AI’s cons?

Jobs: AI is killing jobs and replacing workers. One report estimates that AI is reducing 16,000 jobs per month. However, AI also creates new opportunities and new industries. Another report estimates that between 2025 and 2030, AI and information processing technology trends will displace 9 million jobs while creating 11 million new jobs worldwide. AI also allows people to learn new skills and lowers barriers to innovation and entrepreneurship. Some reports even find that companies investing the most in AI are actually hiring more employees.

Environment: AI technology and the data centers that often house it run on a very large amount of power and water. AI use also increases carbon emissions. However, many researchers and scientists have pushed back on claims that AI is flat-out “bad for the environment” and disagree that outright bans are the best solutions. In fact, some argue that AI can actually make environmental breakthroughs and innovation more efficient.

In terms of water usage, the impact typically depends on where an AI data center is actually located: A data center in a water-stressed region is generally more of a problem than one that is not.

A United Nations report outlining the environmental costs of AI does not conclude that an outright AI ban is necessary. Instead, it calls for more conscious, transparent and responsible AI development. While using less AI can help, regulation and getting AI companies to be more transparent about their energy consumption will have a bigger impact, the experts say. This shows that despite some serious environmental concerns over AI, the conclusion that “AI = bad” is not the most reasonable reaction.

Theft: The “AI is theft!” claim has two versions: First, people argue that “AI steals from artists” because models train on third-party data. However, in a few cases, courts found that training generative AI models on copyrighted works was quintessentially transformative and constituted fair use. Despite those precedents, some newer AI models are now being trained on publicly available or authorized content in collaboration with publishers.

The second is that generative AI art itself is just some stolen copy of someone else’s work. However, that argument oversimplifies how generative AI systems actually work. While some duplicates can occur, AI models don’t typically set out to reproduce an existing work verbatim. Some defenders of AI compare generative AI to students in art school (the ones you can often spot in art museums perched on a bench and sketching a copy of master artwork), because generative AI models also study and learn patterns in order to generate work in similar styles. Even though AI critics reject this comparison, it demonstrates that like human artists, AI systems produce new work influenced by patterns, not by simply copying or stealing the work.

As Kavan Cardoza, a popular AI filmmaker and co-founder of the Phantom X AI production studio, pointed out in a recent interview, we need not look further than Academy Award-winning director Quentin Tarantino as a prime example of this. Tarantino has made a prestigious career out of drawing from the styles of 1970s exploitation cinema, which he grew up admiring. He studied those films’ patterns and learned to create his own in a similar style. Likewise, when people generate AI images, the models they use typically create a unique image, having learned style patterns through training.

This is all to say that AI may be safe to approach with nuance, rather than an adversarial, all-or-nothing mindset. At least, that is how I have approached it and plan to going forward, for now.