Artificial Intelligence Generation: Is It Truly Merely a Tool?
The phrase, "AI is inevitable," stirred in my thoughts for a while, unwelcome and persistent like a catchy tune. I've heard it tossed around in meetings, in emails, and across platforms like Bluesky, LinkedIn, and Discord. It's not a convincing argument, portrayed as either raw hype by AI enthusiasts or a source of fatalism by doomsday prophets. But the philosopher Natasha Bedingfield, in her 2004 wisdom, reminded us that today is where the book begins, and the rest is yet to be written. Nothing, good or bad, is inevitable.
However, another phrase has crept into the shadows, unnoticed amidst the noise: "AI is just a tool. It can be used for good or evil, like any other tool." It's an understandable frame of mind, given our preference for tools in the game development community. We debate whether to use Unity, Unreal, or Godot, whether to go for procedural or hand-crafted content. We even have a whole Tools Summit dedicated to games development software at our annual conference.
But treating AI as 'just a tool' is a limited perspective. It betrays a quiet cynicism that pervades discussions on various topics, like gun control in the United States, for instance. You see, a tool's function is shaped by its form, and nothing, not generative AI, not even a hammer, is merely a tool.
Let's take a look at two household tools: the claw hammer used primarily for pounding nails into wood, and the handgun designed to wound living beings. They are optimized for specific purposes, and we can examine their utility and decide whether they require regulation based on that purpose. But the shape of a hammer is not an efficient means to inflict harm. In comparison, the handgun is. The FBI Crime Statistics survey shows that handguns are the most common weapon used in homicides[1], reflecting their efficiency at causing harm.
Neither a hammer nor a gun is "just" a tool. They are tools designed for a purpose, which we can study, understand, and cast judgement on in terms of safety, utility, and need for regulation. Generative AI is similar. We can use it to deceive through mimicry[2][3][4][5], much like a handgun is designed to cause harm.
Generative AI, for instance, has been used to perpetuate biases in decision-making or educational outcomes[1][3] if it's trained on biased data, just like how a handgun can be used for violence. This highlights the need for responsible AI use, investment in better materials and design to make it safer and more user-friendly, and regular audits to ensure ethical AI practices[5].
In a digital age where reality becomes increasingly blurred, it's essential to acknowledge the ethical complexities of AI and treat it with a nuanced perspective. After all, as Mon Mothma, a character in the Star Wars show Andor, so eloquently put it, "The loss of an objective reality is perhaps the most dangerous. When truth leaves us, when we let it slip away, when it is ripped from our hands, we become vulnerable to the appetite of whatever monster screams the loudest."
[1] "Source for data on handguns being the most common weapon used in homicides": https://ucr.fbi.gov/crime-in-the-u.s/2020/crime-in-the-u.s.-2020/tables/expanded-homicide-data-table-8[2] "Students using AI to cheat on papers": https://www.bbc.com/news/technology-60076015[3] "AI-generated scam calls": https://theconversation.com/artificial-intelligence-is-becoming-an-increasingly-common-tool-for-scammers-161200[4] "DHS report citing secretary Kennedy's unfounded health views with nonexistent studies": https:// processes.hhs.gov/sites/default/files/2021-12/Federal%20Register_12-30%202021.pdf[5] "AI-generated fake movie trailers": https://www.upworthy.com/ai-generated-fake-movie-trailers
- In our digital age, where the lines between reality and artificiality are getting increasingly blurred, it's crucial to approach AI not just as a tool, but with a nuanced understanding of its ethical complexities, much like we consider the impacts and implications of a hammer or a handgun.
- Just as a hammer can be used to build, or a handgun to harm, generative AI can either offer us incredible advancements or perpetuate biases and deception, depending on how it is designed, trained, and used. Therefore, we must invest in responsible AI practices, ensuring it is safe, user-friendly, and aligned with ethical standards, much like we regulate and audit household tools for public safety.
