Even AI is Frightened of AI: Why Human Creativity is Crucial Right Now

I’m trying to be less lecture-y lately, as I fear I’m becoming That Person. And I’m trying not to regurgitate horrific news stories for you or trauma farm or fear monger. We are all plenty stressed. Unfortunately, this post achieves not one part of either of these goals. This is my version of a trigger warning.

What a strange thing it is to be alive as an American right now, and to watch rather helplessly as such vast change transpires each day and yet to have to continue doing what we did the previous day— work, feed the dog, the shepherd kids, clean the house, etc. Every facet of our lives is being defunded, it seems: education, science, medicine, health, the arts, consumer protections, the environment, etc. Also distressing are the extrajudicial deportations often preceded by sudden and sometimes violent kidnappings (particularly and horribly frequent in Massachusetts right now).

I want to explore another threat running alongside the many that stem from our least favorite demented felon/uncle and his posse, and that is the threat of influence that comes from AI. I know many people who love it, and I understand why they do: it is powerful as hell. It can make many of our daily tasks and our jobs easier. It can perform customer service, write term papers, create marketing copy and campaigns, diagnose, teach, research, I could go on forever—and at scale. We don’t have to work so hard if it does the work for us.

And we can use it in other arenas too. More and more people have AI companions, platonic and romantic and, yes, sexual companions that provide understanding and even something that feels like love (I am someone that sees love as a purely human emotion). People use ChatGPT as a therapist, a parent, a teacher. To those who may not know about this yet, I promise you, none of this is fringe stuff. Over 100 million people globally interact with personified AI chatbots, including AI companions. Some estimates are closer to 600 million people. We are well beyond just using AI to keep us from talking to customer service reps.

My kids joke that I suck at identifying AI pictures on instagram.

Four days ago, Zuck said, “I think people are going to want a system that knows them well and that kind of understands them in the way that their feed algorithms do.” He went on to say that he thinks the average person wants to have more friends and connections than they currently do—and that AI friends are a solution. “The average American I think has, it’s fewer than three friends, three people they’d consider friends, and the average person has demand for meaningfully more, I think it’s like 15 friends…For people who don’t have a person who’s a therapist, I think everyone will have an AI.”

Now, if you are like me, you won’t take friendship or therapy advice from a newly bro-ed out dude whose platform has been accused of contributing to horrific mental health outcomes, spreading fake news, mining and exploiting user data, and prioritizing profits over privacy, sanity, and democracy.

Anyhoo, I asked AI (google’s newest bestie) who the largest AI companies are, and here’s what I learned: Microsoft, Nvidia, Alphabet (Google), and Amazon, followed by Apple, Meta, and Tesla are the biggest. Remember who stood proudly by our uncle at his inauguration. But what can Corrupt Uncle do for AI? He can’t even remember that Mattel isn’t a country.

Perhaps least worrisome, he can remove those pesky regulations that Biden began to put in place. But keep reading.

Next I asked AI about the potential of AI to control people, and the response was far too long for this post, so I’ve condensed it here. Note that I am not using quotation marks here, although most of this is cut and pasted from AI. (Ok, the bold font is all mine. But I’m returning the favor after learning that all of my books were used to train AI, as well as the 18 volumes of The Best American Short Stories that I edited, as well as pretty much every book written by everyone I do and don’t know and believe me, I know a huge amount of authors.) Here’s what I learned:

• AI can enhance and automate surveillance technologies, leading to the potential for mass monitoring of individuals and erosion of privacy. (My take: imagine one central source knowing that you attended a protest, got treated for chlamydia, or sheltered your daughter from a physically abusive relationship.)

• Facial recognition, sentiment analysis, and behavioral prediction tools can be used to track individuals, predict their actions, and even identify potential dissenters, potentially chilling free expression and hindering social movements. (Me again: you cannot stand your loudmouth neighbor who collects guns and loves Corrupt Uncle, and you give this person the finger behind her back a couple times, mumbling that you’d like hurt her after she runs over your new rose bush with her Tesla Cybertruck that her billionaire parents bought her. You’re not hugely political. You vote independent. But she is trying your last nerve and you complain about her and her political signs and cybertruck to everyone you know. You can fill in the rest here. Yes, I’m laying it on thick to make a point.)

• The widespread use of AI-driven surveillance can be particularly problematic in authoritarian regimes, where it can be employed to suppress dissent and maintain control. (Me: Imagine ads on facebook showing the violent arrest of people in your state. Now imagine these ads are actually posts. Many of us don’t have to imagine this anymore.)

• AI can be utilized to create and disseminate highly persuasive and targeted propaganda and disinformation campaigns. (Me: see X formerly known as twitter, Fox News, etc.)

• Generative AI, capable of creating realistic text, images, audio, and video (deepfakes), can be employed to manipulate public opinion, incite social division, and even interfere in elections. (Me: see Dumb Uncle as Pope or this and click on any link embedded within. Or the pic below that—shockingly—is not real.)

  

• AI-powered social media bots and troll armies can amplify propaganda, harass individuals, and create a false sense of consensus, making it difficult for people to discern truth from falsehood. (I’ll stop, but after each point, try taking a moment and imagining the impact of each one. Some you won’t have to even imagine.)

• The increasing reliance on AI in critical decision-making processes, like loan approvals, job applications, and criminal justice, raises concerns about fairness, transparency, and accountability.

• If AI algorithms are trained on biased data, they can perpetuate and amplify existing societal inequalities, leading to discriminatory outcomes.

• The "black box" nature of some AI systems, where the decision-making process is opaque, makes it difficult to understand why a particular decision was made, raising concerns about accountability and redress.

• Over-reliance on AI decision-making in various aspects of life could potentially erode human agency and decision-making skills.

• The pervasive influence of AI-driven recommendations and personalized content could limit exposure to diverse perspectives and reinforce echo chambers, potentially impacting independent thought and critical reasoning.

To reiterate, the above is AI’s take on the dangers of AI.

The twin threat of an authoritarian government and the tool that has the power to give them access to the hearts and brains and every one of their citizens is… a big problem.

What to do? This is not about widespread organized resistance, although of course that is necessary right now. I’m writing this to ask you as intelligent, creative human beings to hold tight to your intelligence and creativity.

This means not telling AI your innermost secrets, lest they be used against you in some way.

And continuing to do creative things that start in your own brain and heart, not ideas that ChatGPT or some such has given you— it has enough ideas, and so let’s not give it more.

And continuing to seek joy that stems from human experiences— togetherness, travel, sex, friendship, adventure, theater, film, music, sports events.

And reading and doing word games and going to bookstore events and touring museums and old mansions and giving yourself intellectual experiences that are not offered in three sentences on a social media post or via radio or podcast.

And meeting new people who are different from you, and listening to their viewpoints and not trying to change their minds. My new favorite phrase in the face of someone different from me politically is, “We can disagree.” We can disagree peacefully. It is possible. It is critical.

And being weird. Being messy, impatient, hyper, intuitive, anxious, numbed out, addicted to popcorn and romantic reality TV and literature and my dog—well, that’s me of course, and you should be you, but be the most you you can be, not what some bot tells you to be or think.

It’s Mother’s Day, and as someone who grew up largely motherless (mine passed away when I was 13), I’m not always good at mothering myself, so to all the moms out there— let’s not forget to do this, mother ourselves and preserve our identities, our brains and hearts, our families and world.

Happy Mother’s Day to you all.

Also AI, but come on, doesn’t this look real to you?