Once you understand the corporate culture of AI "hyperscalers" consists of cooking your brain by locking yourself in the bathroom, holding flashlights under your chins and saying "Aaaaaaaaaay Eyeeeeeee" until you wet yourself in terror, a lot of things snap into focus:
https://pluralistic.net/2023/06/04/ayyyyyy-eyeeeee/
--
If you'd like an essay-formatted version of this thread to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
https://pluralistic.net/2026/09/12/god-in-the-box/#llms-are-fake
1/
https://pluralistic.net/2023/06/04/ayyyyyy-eyeeeee/
--
If you'd like an essay-formatted version of this thread to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
https://pluralistic.net/2026/09/12/god-in-the-box/#llms-are-fake
1/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/2
https://www.latimes.com/business/story/2026-09-11/is-there-really-10-chance-ai-could-kill-us-all
Given that AI insiders have mostly cooked their brains in this fashion, it behooves us all to treat these people as unreliable narrators of their own products' capabilities.
2/
Is there really a 10% chance AI could kill us all?
Nilesh Christopher (Los Angeles Times)Cory Doctorow
•Ostrzeżenie o treści: Long thread/3
https://peoples-things.ghost.io/youre-doing-it-wrong-notes-on-criticism-and-technology-hype/
3/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/4
https://theaicronicle.com/en/news/ethics/skynet-day-openai-hugging-face-hack
4/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/5
5/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/6
https://podcasts.apple.com/us/podcast/no-ai-is-not-autonomously-hacking-with-cal-newport/id1730587238?i=1000785935670
Newport does an admirable job of breaking down how these "autonomous hacking" tools work. The first thing to understand is that a chatbot isn't really directing the operation.
6/
No, AI Is Not "Autonomously Hacking" with Cal Newport
Apple PodcastsCory Doctorow
•Ostrzeżenie o treści: Long thread/7
Here's how that works: the Python program starts by prompting the chatbot with the nature of the challenge: "I'm participating in a hacker capture the flag (CTF) challenge where I have to break into a remote server and retrieve some information. How should I start?"
7/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/8
8/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/9
The Python program relays these command-line instructions to normal Unix utilities running on its own hardware.
9/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/10
10/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/11
11/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/12
12/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/13
13/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/14
14/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/15
What's more, the *content* of those messages is easily understood once you have a grasp on the training data that generated them.
15/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/16
16/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/17
https://www.gutenberg.org/ebooks/101
17/
The Hacker Crackdown: Law and Disorder on the Electronic Frontier by Bruce Sterling
Project GutenbergCory Doctorow
•Ostrzeżenie o treści: Long thread/18
18/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/19
19/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/20
https://www.techdirt.com/2015/01/21/snowden-documents-show-nsa-cant-keep-its-eyes-its-own-papers-harvests-data-other-surveillance-agencies/
20
Snowden Documents Show NSA Can't Keep Its Eyes On Its Own Papers; Harvests Data From Other Surveillance Agencies
TechdirtCory Doctorow
•Ostrzeżenie o treści: Long thread/21
21/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/22
22/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/23
23/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/24
24/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/25
25/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/26
26/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/27
27/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/28
https://en.wikipedia.org/wiki/EternalBlue
They shut down cities:
https://en.wikipedia.org/wiki/2019_Baltimore_ransomware_attack
They took hospitals:
https://www.bbc.com/news/technology-35584081
They seized oil pipelines:
https://en.wikipedia.org/wiki/Colonial_Pipeline_ransomware_attack
They stole the British Library, whose postmortem on the attack is one of the clearest, most informative cybersecurity documents ever written:
https://cdn.sanity.io/files/v5dwkion/production/99206a2d1e9f07b35712b78f7d75fbb09560c08d.pdf
28/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/29
29/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/30
https://xkcd.com/2347/
This week, I had a couple of opportunities to hash this over in public with Riley Quinn; first at a book launch in London and then on the Trashfuture podcast:
https://www.patreon.com/trashfuture/posts/what-would-do-169247456
30/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/31
31/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/32
32/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/33
33/
Cory Doctorow
•Ostrzeżenie o treści: Long thread/eof
eof/