commit 3f5546735582bbf1fc4cd0472bb1e5d4a9c94d5b
parent bff8691082a747376c0c9c1d6a3580e42e5252df
Author: Eamon Caddigan <eamon.caddigan@gmail.com>
Date: Tue, 14 Jul 2026 15:35:22 -0700
Fix a typo and add a footnote
A couple folks kindly pointed out a grammar error. I also added a link
that I found shortly after publishing that I've wanted to include here
since then.
Diffstat:
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/content/posts/ai-in-2026/index.md b/content/posts/ai-in-2026/index.md
@@ -14,13 +14,13 @@ I was proposing my doctoral dissertation fifteen years ago when neural network m
People have lost their minds over a _neat trick_. There was a period, hopefully nearing its end finally, when it was _common_ for otherwise sensible people to think that next-word-generating machines were close to achieving human-like consciousness. Experts in cognition and language have continually explained why this won't happen, but that's a mania for you. People still call this whole suite of technologies "artificial intelligence," so that's what I'll do here.
-What does it feel like to work in tech right now and exist outside this spell? Few of my contemporaries want hear about it, but to preserve my sanity I will imagine that someday, somebody might.
+What does it feel like to work in tech right now and exist outside this spell? Few of my contemporaries want to hear about it, but to preserve my sanity I will imagine that someday, somebody might.
The largest tech companies---which already comprise an uncomfortably large portion of the economy (in spite of the fact that they produce none of the necessities of life)---have bet their fortunes on AI. Actual innovation in neural networks themselves has slowed as the companies and investors focus on building larger models, requiring tremendous amounts of training data and energy to create.
It's worth highlighting that explicit climate denialism is well out of vogue here in 2026. Most people profess belief in the scientific consensus that society's dependence on fossil fuels is warming the climate. This stops few of them from using a technology that is directly increasing society's thirst for oil. When people like me argue that this technology is unfit for adoption on the basis of its energy requirements alone, AI users will simply ignore anything that we have to say.
-This is the template for a common pattern. We can point out all of the harms being done by the technology _right now_, and we can speak to the additional harms that will be done if things continue like this, but our concerns are spurned or ignored. AI users seem untroubled by---or unaware of---the dissonance, because the richest individuals and corporations have decided that the future must be built on this specific approach to this specific technology that they own.
+This is the template for a common pattern. We can point out all of the harms being done by the technology _right now_[^harms], and we can speak to the additional harms that will be done if things continue like this, but our concerns are spurned or ignored. AI users seem untroubled by---or unaware of---the dissonance, because the richest individuals and corporations have decided that the future must be built on this specific approach to this specific technology that they own.
Some people find chatbots useful and I can't argue with their opinion, but nobody is doing anything with them that remotely justifies their socialized costs. I'm compelled to note that I've been a machine learning practitioner for 20 years, and was excited about last decade's neural network developments, but this train has gone off the rails.
@@ -31,3 +31,5 @@ Something seems to be lost on my peers today: **it's still easy to not use AI**.
If any but the most optimistic prognostications about this technology come to pass, chroniclers will wonder why more people didn't do that. I'm baffled too! My emotional state in this moment recalls that from late 2002, during the lead-up to the US invasion of Iraq. It seemed obvious then that the people leading the nation were lying to justify war, and that nothing good would come of it (for anyone but a small handful of rich people anyway), yet the media and much of the population treated critics with absolute contempt. Being anti-AI feels the same as being anti-war. The salient differences between then and now are that the bloodshed will be harder to attribute (how many people are killed by one tenth of one degree of global warming, and what proportion of that temperature rise comes from one specific gas-burning data center?), and that more of my actual peers---including dear friends---are on the wrong side of this. The familiarity stings, but the pattern is repeating itself because nothing has fundamentally changed about us since then.
I've struggled to write something that would persuade my colleagues and friends to ditch AI and affirm their avowed beliefs about climate change, the trustworthiness of megacorporations, and our right to live and work with dignity. I accept that I'm not up to that task. So I wrote this instead for anyone who someday wonders what the hell is wrong with us. I can't ask that reader to forgive us, only to understand that we're not bad people. We're regular people living in a situation that rewards bad decisions and punishes good ones.
+
+[^harms]: I found [To Gen or Not To Gen: The Ethical Use of Generative AI](https://blog.johanneslink.net/2025/11/04/to-gen-or-not-to-gen/), by Johannes Link and Jakob Schnell, after I first published this. I don't know whether I'd have written this otherwise---their article carefully outlines the harms perpetuated by this technology. [The AI Con](https://thecon.ai/) (Bender & Hanna, 2025) is also recommended reading.
