Sign in

Very Sane AI Newsletter

@verysane.ai
1K followers 0 following 25 posts

Ideas in AI, preferably in English. Mostly what and why, a little how.

PostsRepliesMedia
Very Sane AI Newsletter @verysane.ai · 08/08/2026
Who deserves the most style points for how their LLM commits felonies? (it's openai) verysane.substack.com/p/felonybenc...
verysane.substack.com
FelonyBench Style Points
There are many benchmarks we use to rank LLMs, but “how many felonies has it committed” has recently become popular.
312334
Reposted by Very Sane AI Newsletter
Evan Goer @evangoer.bsky.social · 15/06/2026
when your leadership is too off-putting and weird to hire & retain normie attorneys and PR reps
1122
Very Sane AI Newsletter @verysane.ai · 14/06/2026
Did Anthropic ask for this? www.verysane.ai/p/did-anthro...
verysane.ai
Did Anthropic Ask For This?
This past Friday, the US Government issued an export control directive that prohibits Anthropic from giving foreign nationals access to Claude Fable or Claude Mythos, their latest models.
0653
Very Sane AI Newsletter @verysane.ai · 11/04/2026
"sort of a drunk Sisyphus situation [...] Orthogonality Thesis [..] superintelligence with a randomly generated utility function [...] Yudkowsky [...] If the übermenschen align the AI, who aligns the übermenschen?" tldr AI doom is wrong www.verysane.ai/p/counting-a...
verysane.ai
Counting Arguments and AI
A “counting argument” is a style of argument common among creationists, who argue that the theory of evolution cannot be true and therefore humans (and usually animals too) were made in basically their present form by God.
37612
Very Sane AI Newsletter @verysane.ai · 11/04/2026
There's another one of these going out tomorrow morning and it might be a while after it publishes before it gets posted here so bear with me
020
Very Sane AI Newsletter @verysane.ai · 29/03/2026
"Luddism does not deserve to be rehabilitated. It was a medieval throwback, reactionary and primitive, a pre-Marxist labor convulsion closer in spirit to the Khmer Rouge’s fantasies of agrarian restoration than to the universalist solidarity of Eugene Debs." open.substack.com/pub/verysane...
open.substack.com
Against the Luddites
The rehabilitation of Luddism is a vice signal.
826070
Very Sane AI Newsletter @verysane.ai · 26/03/2026
"I hope that we can, at some point, have some reasonable regulations on AI, as we do concerning banks, wiretaps, and dangerous chemicals. I am somewhat pessimistic about legislation we have seen in the past, and legislation we are seeing now." www.verysane.ai/p/some-rough...
verysane.ai
Some Rough Notes on AI Policy
I hope that we can, at some point, have some reasonable regulations on AI, as we do concerning banks, wiretaps, and dangerous chemicals.
1284
Very Sane AI Newsletter @verysane.ai · 16/03/2026
"Stochastic Parrots" is wrong
verysane.ai
Polly Wants a Better Argument
The “Stochastic Parrot” Argument is Both Wrong and Actively Harmful
918051
Very Sane AI Newsletter @verysane.ai · 09/03/2026
If LLM consciousness is interesting to fight enough it should be interesting enough to read seriously about, right?
verysane.ai
Might An LLM Be Conscious?
In short, this depends on what you think that means, whether you think it’s possible in principle, and what you think would be evidence of it.
107612
Very Sane AI Newsletter @verysane.ai · 27/02/2026
this is about the government using AI for surveillance a somewhat rushed post about Anthropic telling the Secretary of War no, Anthropic's history with the government, etc.
open.substack.com
Claude's Custody Hearing
Ideas in AI, preferably in English. Mostly what and why, a little how.
14713
Very Sane AI Newsletter @verysane.ai · 18/02/2026
That LLMs understand natural language as well as they do should dramatically change our understanding of the problem of 'making AI do what we want it to do'. www.verysane.ai/p/alignment-...
verysane.ai
Alignment Is Proven To Be Tractable
That LLMs understand natural language as well as they do should dramatically change our understanding of the problem.
6847
Very Sane AI Newsletter @verysane.ai · 14/02/2026
it is probably not a good idea to put GPUs in space www.verysane.ai/p/should-we-...
verysane.ai
Should We Put GPUs In Space?
Right now? No. In the near future? Maybe. Probably not, though.
2336
Very Sane AI Newsletter @verysane.ai · 12/02/2026
My take on Searle's Chinese Room, which is terminally engineer brained in that my entire argument is "okay, how would you actually build it, though" www.verysane.ai/p/building-t...
verysane.ai
Building the Chinese Room
Suppose also that after a while I get so good at following the instructions for manipulating the Chinese symbols and the programmers get so good at writing the programs that from the external point of...
7627
Very Sane AI Newsletter @verysane.ai · 08/11/2025
An attempt to analyze recent legal filings about AI enabling user suicides, figure out what caused them (OpenAI being irresponsible, mostly), and figure out how to deal with this, and any similar problems in the future. www.verysane.ai/p/ai-and-sui...
verysane.ai
AI and Suicide
[Warning: Extensive discussion of suicide.
05912
Very Sane AI Newsletter @verysane.ai · 14/08/2025
There is some confusion about whether or not we understand LLMs. The answer is yes and no, but mostly no. It's a complicated enough question that it seemed like it needed an article. www.verysane.ai/p/do-we-unde...
verysane.ai
Do we understand how neural networks work?
Yes and no.
1330791
Very Sane AI Newsletter @verysane.ai · 12/08/2025
"AI 2027" argues that AI will reach roughly human level in roughly 2027. This just happens to be right about when OpenAI would expect to start to run out of money. I argue that this is not a coincidence at all, and its predictions are all wrong. www.verysane.ai/p/agi-probab...
verysane.ai
AGI: Probably Not 2027
AI 2027 is a web site that might be described as a paper, manifesto or thesis.
1841286
Very Sane AI Newsletter @verysane.ai · 15/07/2025
What Makes AI "Generative"? It's Not What You Think I can just tell you actually: it's how large the output space is, or equivalently, how many possible outputs there are. Bigger output spaces are just fundamentally different. www.verysane.ai/p/what-makes...
verysane.ai
What Makes AI "Generative"?
To "generate" is to create, either from nothing (The Book of Genesis), or from very different or relatively inactive materials (an electrical generator, generating offspring).
1163
Very Sane AI Newsletter @verysane.ai · 01/07/2025
Some Thoughts On The Platonic Representation Hypothesis, or, How We're Rediscovering Platonism By Doing Statistics or, How AI Training Is Actually Chaining Up Some Guy In A Cave, or, Bait for Philosophy Majors www.verysane.ai/p/some-thoug...
0273
Very Sane AI Newsletter @verysane.ai · 09/06/2025
The most well-known statistic about AI water use is a lie. This makes it frustrating to talk about AI and the environment, and this is a long deep dive on that specific point. www.verysane.ai/p/the-bigges...
verysane.ai
The Biggest Statistic About AI Water Use Is A Lie
How did it become the main story?
1446
Reposted by Very Sane AI Newsletter
Very Sane AI Newsletter @verysane.ai · 07/06/2025
Pushed previously as a notes file, this is a short list of ideas which have, or are, setting the agenda for AI, presented entirely by quoting the relevant person. www.verysane.ai/p/ai-history...
verysane.ai
Key AI Quotes
Each of these presents the clearest, earliest, or most-cited statement of a specific idea in AI.
093
Very Sane AI Newsletter @verysane.ai · 07/06/2025
Pushed previously as a notes file, this is a short list of ideas which have, or are, setting the agenda for AI, presented entirely by quoting the relevant person. www.verysane.ai/p/ai-history...
verysane.ai
Key AI Quotes
Each of these presents the clearest, earliest, or most-cited statement of a specific idea in AI.
093
Very Sane AI Newsletter @verysane.ai · 04/06/2025
I couldn't find a high-level overview of what AI is and how we got where we are, so I wrote one. www.verysane.ai/p/what-is-ai
verysane.ai
What Is AI?
This is an important question because AI is, currently, important.
15914