By Madelyn Perry
In the spirit of transparency, I am an AI skeptic, and this is in large part because I do not think that AI, although an umbrella term, is yet capable of working within that same spirit. In the words of Dr. McCoy, “transparency takes words, and words take work, care, and time.” Because of what I would best describe as The AI Race, or the eagerness with which both big and small tech companies are putting out their own Generative AI models and even updates to already existing models despite not having all the kinks (like the safety measures on Character.AI that could have directed Sewell Setzer III to a suicide hotline) worked out, it is clear that these models do not need to be forthcoming or honest to be considered successful. Perhaps that is their most human quality. And as someone who not only has to do a lot of writing but also enjoys the process, I naturally have a host of concerns regarding the future of creative professions, not the least of which is how I will guide my future students on not only how to use AI ethically, but what ethical AI use even looks like. However, with all of AI’s intended purposes, I believe that its most inspiring effect will be something unintended, something that did not require any data or programming. Because AI has taken the world by storm, the response to its existence has been equally forceful, and the strength behind the human response to AI demonstrates the responsibility consumers have to AI’s success.
While the advent of AI is often compared to that of Google (in that they incited similar fears), AI’s reach is far wider, making it widely accessible, but inescapable to those not wishing to partake in it. In fact, Google’s own form of AI, called Gemini, offers a summary response to any question asked, and does so at the very top of the page. This means that nearly everyone who has internet access and has Googled something within the past two years has some degree of familiarity with AI. But AI is not just an advanced search engine. It can generate art, music, writing, as well as solve complex math problems. So, the degree to which AI is causing a disruption in ordinary tasks, such as summarizing and writing emails, and is encroaching on more creative endeavors, means that now more than ever before there is a call to put policies and regulations into place. For instance, fake news is no new threat. It has especially been a hot topic for the past three elections. The difference this time around has been, again, the farther reach, the increased scale of production, and as John Herrman points out in his article, “Taylor Swift and the Power of AI Backlash,” the lies that are “stubbornly immune to correction because nobody thought they were strictly true in the first place.” This indicates that individuals very experienced with social media are becoming skilled in differentiating between AI and non-AI. But evolution of media awareness does not excuse the generation of images that are created with harmful intent. The intent to harass may affect less people than the blatant spread of lies, but it is harm done nonetheless. While I admit that my skepticism is directly underpinning my thought process, the explosion of AI and its subsequent rocky reception means that now, more than ever before, people are attentive to unethical tech use. This creates a breeding ground for positive change, things like safety features, banning the use of anyone’s image or likeness without consent, or the generation of images based on someone else’s art or other creative work unconsented or uncredited.
The scale at which those issues are happening demands an equal reaction, one that can only come from humans: lawmakers, judges, attorneys, folks in Congress, and folks like Dr. Joy Buolamwini who did not realize she had a right to be included in the AI conversation and a seat at the table where it was happening until she, a computer scientist herself, found herself “ex-coded,” something she describes as being excluded from the target audience that a program is designed to reach, when the code she was using did not recognize her Black skinned face as human, but did so when she wore a white mask. That she, someone who actively wished to escape society through technology, later felt so compelled to found the Algorithmic Justice League, an organization that works to inform people on AI’s inherent biases in an attempt to correct the way that technology best serves privileged demographics, is a testament to the active rather than passive role human beings can be taking to guide the development of AI. Many people, from both ends of the spectrum (from “AI is the future” to “AI is the undoing of our civilization) have described AI as inevitable, as if to suggest that this is just something that has to happen to us as a society, as if to suggest that efforts the ordinary person might make to direct technology to serve ordinary needs are futile, and that tech’s purposes instead need to be decided by the billionaires who create it. Incidentally, this is the exact feeling that big corporations aim to produce, more so I would argue, than their physical products themselves. To see people then rising up as specialized AI lawyers for instance, represents the passion that AI has awakened in people to become figures of social justice, which is something that was likely unforeseen, but inspiring nonetheless.
And, while it may be a lofty claim to suggest that AI, a tool that threatens to make creative professions obsolete, could go so far as to catalyze something like a Renaissance, artists will undoubtedly have to change the way they create, whether in an attempt to prove their work is not AI generated or instead to leave a message along the lines of “Look what I can do that AI cannot.” This shift, I am willing to predict, will years from now mark a time of significant social upheaval. As these are the conditions that creative movements traffic in, there is a lot to look forward to in the art scene. AI could help to produce art, or it could create a drive or sense of intrinsic motivation (even anger) that many artists might thrive in, feeling inspired not necessarily by AI itself, but the recognition of their own human capabilities and sensitivities, something that AI has yet to master.
Ultimately, AI is only as inevitable as we consumers allow it to be, and for those like myself who struggle to not fall into a total doomsday spiral, this is enough of a reminder that AI is not inherently a bad faith actor but is instead a reflection of our society’s values. AI, because it is not yet sentient, cannot and is not asking us to use it, but it still poses us with the question of how we could use it. However, to first determine that, we need to determine how we should not use it. That said, the movement to regulate Big Tech is perhaps the largest step we can take to direct AI towards pursuits like cancer detection or translators or tutors for non-native speakers. This cannot be done without human involvement. AI cannot regulate itself. Although AI has been marketed as something that will do our labor for us (for the better part of 70 years), we have since discovered that we affect AI as much as it affects us, and that the labor it does do for us is only as qualitative as our continual molding of it, thus AI inspires me, and hopefully inspires all of us, to hold ourselves accountable for that relationship.