I’ve been participating in organizing around stopping an AI project in my community (namely a data center collaboration between the University of Michigan and Los Alamos National Labs), and I’ve been thinking a lot about how merely stopping these initiatives seems insufficient. I’ve been thinking about what would be sufficient, and I’ve come to appreciate, through organizing, that as much as “destroying AI” means uprooting AI initiatives, it means building counter-structures that render AI interventions totally incomprehensible to us.

I think I wrote in my techno-optimism piece that AI seems to thrive opportunistically, not totally unlike a weed in a garden. Maybe I’ve been thinking about this because I’ve been doing some gardening lately, but stay with me; weeds take root where the environment has resources to extract, and where other things are not yet growing. Where too many other things are thriving, weeds tend to have much more trouble taking root. You can spend all day uprooting weeds, but frustratingly the best thing you can do is cultivate a healthy environment and support the life that you want to thrive; healthy plants will deny weeds access passively, mostly on their own.

We can organize to stop data center proposals; we can demand our cities cancel surveillance contracts with Flock; we can shout down the commencement speakers who try to browbeat us into submission under the idea that “AI is inevitable”. I do some of those things when I can, and we must do all of those things across the board. But that doesn’t make our community any less hospitable an environment for “AI solutionism” to take root and thrive as soon as the winds change and we falter. You could blow up Flock HQ and in 5 years there’ll still be surveillance companies offering their services to police, because there is a hospitable environment for such a business and we give them space to take root, like weeds.

After reading Health Communism by Artie Vierkant and Beatrice Adler-Bolton, I spent a lot of time thinking about the premise they laid out for the reader - a society that mobilized the bulk of its resources to care for the sick, that didn’t make that care conditional, that moved mountains for the vulnerable rather than leveraging people’s vulnerability to stoke desperation and anxiety. What would a society look like that denies the use of threats of homelessness and destitution to coerce people? What would our lives be like if we structured our society to mobilize care for all of us?

I believe that we will not be able to stop Flock just by canceling contracts across the country. I believe that we will not be able to destroy AI just by stopping data center proposals. I think it’s important to make the costs of pursuing AI exceptionally high, but it’s also important to render the potential benefits of AI enterprises as close to nil as possible. There has to be no room for these weeds to take root and proliferate, or we’ll constantly be chasing data center proposals, browbeating city counselors for signing surveillance contracts, etc… and as soon as we’re exhausted, they’ll overrun us.

Destroying AI must include building counter-structures and nurturing a healthy, thriving social landscape that denies AI projects access to us in the first place. AI solutions like therapy & medical chatbots find space to thrive because of all the gaps in medical care we’ve normalized; we must make these interventions totally inscrutable in a future where care is always available, and people’s needs are not constantly being means-tested and scrutinized. Surveillance interventions and police collaborations like Flock thrive because we subscribe to this police state in which we live, where surveillance and violence are natural extensions of the logic motivating state violence; we must build a future where the impulse for the state to commit violence or to intrude on our privacy registers as bewildering and alienating.

Uprooting these AI projects is what we do to address a past mistake in leaving space for these things to grow in the first place. We must both uproot these projects and we must work to build a future where these kinds of proposals simply make no sense to us, because we have reoriented everything we have to take care of one another earnestly and patiently.

I don’t have a blueprint for this; it’s just a thought that has been lingering at the periphery for a while that I’ve wanted to get out of my head. I worry that the University of Michigan is not motivated by exceptional forces in their exceptional project to build this billion-dollar military facility with the lab that built the nuclear bomb. I don’t believe healthcare, education, and a myriad of other industries are reacting to exceptional influences when they embrace AI; I think they’re being overrun by conditions that are conducive to the invasive weediness of AI, and I think in every setting we need to think about what counter-structures we need to nurture and sustain to make these interventions make no more sense than coupons for things that are already free, or a way to fast-forward through leisure time. I don’t know if we can destroy AI just by chasing around the sprouting invasive plants; I think we need to grow and foster other things that make weeds struggle.