happy kirkiversary to all who celebrate

Anthropic Is Building a Predictive Surveillance System to Monitor Activists

Job postings and interviews with senior security officials at Anthropic show that the frontier AI lab is building out an extensive monitoring system to keep tabs on activists who oppose the rapid development of artificial intelligence.

In addition to monitoring activists in the vicinity of Anthropic executives and keeping tabs on protests near physical Anthropic assets, the firm is also implementing a “pre-crime” approach, attempting to predict incidents before they happen. In some cases, that also means reporting suspects to police before a crime occurs. Anthropic did not respond to the Prospect’s request for comment.

Apparently they can also sick the police on you on a trust-me-bro basis:

Last month, The San Francisco Standard reported that Anthropic had reported a man to San Francisco police for telling Claude that he had bought an AR-15 semiautomatic rifle and had CEO Dario Amodei “in his sights.” When the Standard contacted the man in question, he told the newspaper he was “just fucking around.”

But while Anthropic was fast to call the cops on a frustrated Claude user, the Standard also reported a key detail: Anthropic refused to show police the actual messages, citing Anthropic’s internal policy. In short, Anthropic reported a user for in-platform speech, and then refused to provide police with evidence of actual wrongdoing.

image shamelessly stolen from https://bsky.app/profile/timhenke.bsky.social/post/3mv5puo47ns2z

Terry Tao talks about how he used to try to cooperate with the AI industry to achieve a positive outcome, but now he finally sees their true colors. Link

During this event, OpenAI requested an interview concerning my vision of the future of AI and mathematics. I accepted, and spoke with them for perhaps an hour. I had done similar interviews in various venues, and I assumed that, as with these other cases, they would eventually post the entire interview online, which talked about both the possibilities and risks of AI much as I have done in these other interviews. As it turned out, they only used a few snippets of that interview for that infamous advertisement instead. In retrospect, I should have pushed back harder on their decision; but I decided at the time that even a selective release of my commentary would help raise awareness of the potential for AI, and in particular on the possibility of the “best of both worlds”.

Since then, the situation has deterioriated markedly. Many of the people in the industry that shared my views have left or become sidelined, with most major tech companies now increasingly focused on the race to develop extremely powerful, autonomous AI technologies regardless of their actual value to society. The current drama surrounding the Navier-Stokes global regularity problem is the most dramatic and visible instance of this, but there have been multiple other such examples, and much of my commentary in the last few months has been aimed that the increasingly severe divergence between the current objectives of the AI industry, and of mathematics in general.

Much respect to artists for seeing all this coming from the very beginning, and holding the line.

Looks like there might be a bit of an openai vs anthropic dick waving competition over mathematical proofs. I haven’t had time to digest all of this yet because it is a bit hard for me to read, but there are a few interesting nuggets in there such as making it clear that the ai firms treat this as a marketing exercise (no surprises there). They’re also apparently trying to obscure the amount of human assistance involved (again, no surprises).

Levent had been told by Sebastien “very little human input” had been used. This turned out not to be true. Over the course of the call, as members of their team sent Sebastien corrections and details over their internal chat, it emerged that an entire team had been working on the problem, that this was one of a number of things that was tried, that work had started on the unforced problem, that the team first set the model on easier problems, including Euler, that even the prompt that had been shown to me had been written by prompting Codex, and that an insane amount of compute had been used.

The main statement is here: https://cims.nyu.edu/~tristanb/statement.pdf

Associated stuff linked from here: https://mastodon.social/@tristanbuckmaster/117233413705701198

Do note that the author made heavy use of llm assistance for their work, and does consider it to be the future of developing mathematical proofs… this isn’t human vs ai.

Tesla’s new self-driving taxi seems to be entirely made of red flags and those horror movie moments where you can’t understand why anyone would go in there, because they’re clearly going to die.

https://archive.is/2026.09.05-135619/https://www.nbcnews.com/tech/tech-news/tesla-cybercab-no-steering-wheel-emergency-plan-push-musk-austin-rcna595741

Highlights:

• No door handles. Good luck getting out instead of burning to death and/or drowning!
• Doors might not unlock if there are power issues, eg. Bad things happened in a crash.
• No manual controls! First responders will have to push it out of the way.
• Unless there’s a power issue, because the parking brake will engage if there’s no power, so it’ll need to be barged by another vehicle, I guess?
• Or you push it faster than 5mph, because the brake will engage for safety reasons.
• Or if it is in the wrong gear or mode, because back-driving the motor can cause electrical safety issues or overheating, which will presumably increase the risk of people trapped inside being burned to death.
• The vehicle is silent and may move without warning. You can’t tell if it is unpowered. First responders are instructed to “cage” the vehicle before trying to interact with it. Otherwise I guess it might just main or kill them by surprise.
• Not all high voltage components are labelled. Exposed bits in a crash might just kill you if you don’t have suitable ppe, and you can’t tell which bits those are, lol.

The guide advises first responders to contact the company via the car’s communications system or phone for most emergencies, and says that the Tesla can assist emergency personnel by moving the car remotely, helping provide access to on-board controls or putting the car in neutral so that it can be pushed.

If Tesla cannot be reached by phone, the manual advises first responders to use an email address to learn more about how to move the Tesla, though it warns, “This email is not monitored for emergency requests.”

🤡

Welcome to the future, where first responders might just opt to watch you die because they’re not certified to rescue you from your autonomous death trap, because there’s no point in you all dying.

H/t to violet blue. https://mastodon.social/@violetblue/117237691936474802

without going into the merits of the specific case, I find it highly amusing that OpenAI fucked up their big Navier-Stokes announcement which is now more about how they stole research from honest mathematicians rather than how awesome their AI is

New (to me) lore from https://www.theargumentmag.com/p/god-is-back-on-the-timeline

This all kicked off last year when religious Substacker Ethan Muse wrote a heavily sourced essay arguing that the Miracle of the Sun at Fatima — a 1917 incident where large crowds of people reported seeing the sun dance and emit radiant colors — was a genuine miracle, unexplainable by any other method other than divine intervention. This inspired rationalist blogger Scott Alexander to write a more skeptical post titled “The Fatima Sun Miracle: Much More Than You Wanted To Know,” which led to responses and counter-responses.

The discourse continued into 2026 when other bloggers began jumping into the fray and Alexander began investigating similar sun miracles in Buddhism. It seems to have wrapped up with Alexander’s recent pilgrimage to a modern-day holy site where people still claim to see sun miracles to this day.

lol pilgramage what. is he converting?

Greg Egan reports:

Open Letter about the Mathathon

This open letter calling for the cancellation of the Caltech “Mathathon”, has received 471* signatures so far. [Only members of the Caltech community or mathematicians at PhD student level or beyond are eligible to sign.]

* Now 771 signatures.

“On October 30th, 2026, Caltech will begin hosting the Caltech Mathathon, a math hackathon in which participants are encouraged to solve open research problems by prompting LLMs. Two prominent AI companies, Anthropic and OpenAI, will supply participants with 2 million dollars in AI credits. This event is likely to have destructive impacts for the mathematical community.”

https://docs.google.com/forms/d/e/1FAIpQLSf6dBeDwjRP4UkchGSpLwLQjiG8qLX5_wW_3-xoC4dImTPp6g/viewform

https://www.theregister.com/ai-and-ml/2026/09/09/ai-models-dont-kill-people-people-kill-people/5295368

At the same time, some of that harm may be balanced on a statistical level by lives saved through AI tech.

[citation needed]

It was only a matter of time. Enlightened centrist Brodie Robertson is now rubbing shoulders with DHH on his podcast. I guess after making nice with the likes of Andreas Kling, Vaxry and Lunduke it was only a small step to invite DHH as well.

Another oped from StudyHacks Cal Newport:

Opinion: How Scared Should We Be of A.I. Right Now? | NYT

The time has come to push back. Once we understand the key ideas of Rationalism, we can begin to filter them out. Be very wary, for example, when A.I. leaders or online commentators begin talking in sweeping generalities about the future of the technology, such as stating with confidence that many jobs will soon be automated or that their language models will soon become sentient beings intent on escaping their creators’ control. (No, OpenAI’s autonomous agents conducting hacks doesn’t mean that A.I. is rising up against its human overlords.)

The correct reply is to promptly turn the conversation back to the possibilities, harms and viability of the products that actually exist. To extend an analogy I’ve made before, if Ford automobiles were spontaneously catching on fire, we wouldn’t tolerate the chief executive’s changing the subject to discuss our general lack of preparedness for a future with flying cars. We’d demand that the company address the acute problem happening right now.

(Hey remember when there was a guy that actually blew the whistle on exploding cars, ran for president? “We” tolerated exploding cars and people dying for years, until someone actually stepped up and forced the issue. How’s the country been treating that guy?)

It’s not surprising that so many Americans feel negatively about A.I. at the moment, but these sentiments aren’t inevitable. Our current crop of L.L.M.-powered tools are exciting — and worrisome, and promising, and frustrating, all at the same time. With proper direction, they’ll hopefully make our lives better. But they don’t have to be terrifying. We thought Silicon Valley was trying to warn us of an unavoidable danger, but maybe they were just being weird.

wonder if such a moderated, watered down position in a large publication ends up sending more eyeballs to Yudkowsky than it does to innoculate the general public, could end up a net negative.

Another Anthropic employee has quit for reasons typical of an Anthropic employee

The latest tech bro attack on public education is Alpha School for teenagers. One reporter thinks that if we take the accounts claiming to be students are real then:

…we have to take semi-seriously the question of whether Alpha School is a cult? A set of indoctrination camps for the airport-book class? What the hell is happening here?

A focus on judging people by biometrics, cult criticism, and assertion that LLMs can replace teachers. Someone on Bsky replies:

The cult of Alpha school is strong in Miami-where the Miami tech/finance/crypto bro set is just generally dumber (or less intellectual, less cultured, less educated) and far less ethical than in other parts of the country (not that silicon valley tech bros are awesome, but the worst ones move here)

Hereticon got started in Miami.

wtf is deal of that segyges person. now they’re convinced that openai will solve room temperature superconductors in silico real soon, because altman vagueposted about it

and this all based on lk-99 fraud case, nevermind that there’s no high temperature superconductivity theory, or how prohibitively computationally expensive computational chemistry is, and i guess it’s not accurate enough anyway, this all with people that do computational chemistry coming in their thread trying to explain this, then posts this

Crosspost, hat-tip to @pnwpixel@programming.dev for finding this. Omarchy announces another patron: DigitalOcean. I will wait for DO to explicitly confirm/deny this, but if this is true then I’m going to be moving my entire small-business footprint.