StatsAreFun
Social influence and behavioral contagion experiment with bots on Reddit
Code-Mixing on Sesame Street: Dawn of the Adversarial Polyglots
Re:On the Navier–Stokes Millennium Prize Problem
on Tuesday September 08, 2026 @04:16PM (#49616382)
Sure, can do.
Re:On the Navier–Stokes Millennium Prize Problem
on Tuesday September 08, 2026 @04:11PM (#49616304)
Publicly available models are pretty good but seemingly cannot compete with these Astra++ internal-only models.
Re:On the Navier–Stokes Millennium Prize Problem
on Tuesday September 08, 2026 @03:49PM (#49615970)
Can't help but shake an unsettling feeling about all this, frankly. I engage in some limited mathematical research and will often use any one of the latest frontier models to check some ideas. Lately, only the OpenAI models have been giving me a temporary message that says something like (paraphrasing from memory), "We're thinking extra hard about your request before we answer. You can choose another model to answer now or click here to learn more about why." When I click to read why it's doing this "extra thinking", the help page says that for cybersecurity and biosecurity-related information, it will review the answer and could refuse.
Now, keep in mind, I'm only asking strictly pure mathematical questions - nothing at all related to cyber or protein creation or biohacking or anything like that... And, like I said, only the OpenAI models are doing this. To be fair, all of the prompts have always eventually returned a satisfactory answer, as far as I can tell, and haven't used a weaker model to answer them. Maybe? I dunno, it has just struck me as odd every time it has given me that message to pure math prompts.
Despite New Updates, Flock's Creepy Cameras Remain Major Civil Liberties Threat
What shifts threshold distributions in social contagions?
Re:No duty to warn if products contain risks not identified at federal...
on Wednesday August 12, 2026 @11:18AM (#49273779)
This was a court case that went to the SCOTUS. Official reference here: https://supreme.justia.com/cases/federal/us/609/24-1068/
No duty to warn if products contain risks not identified at federal level
Re:Oracle bans AI-generated code from OpenJDK
on Friday August 07, 2026 @05:47PM (#49216605)
From the page https://openjdk.org/legal/ai -
"6 Is it okay to continue using the spell-checking, grammar-checking, auto-completion, and refactoring features in my editor or IDE?
Yes, so long as they are not based on large language models or similar deep-learning systems."
The International Lunar Reference System
Lost in Context: Addressing Context Anxiety in Large Language Models
An Early Warning of Emerging Biosecurity Risks in Frontier LLMs
The Direct and Indirect Effects of Genetics and Education
The Right to Privacy (Harvard Law Review, 1890)
Honey Bee Colony Monitoring via Audio IoT Sensors, Tensorgrams and RNNs
Anthropomorphism in Children's Interactions with LLM Chatbots
Flock Credibility Lost as It Repeatedly Lies to City Councils, Police, & Public
Re:ACLU: Do Automatic License Plate Readers violate privacy under the ...
on Thursday June 18, 2026 @11:44AM (#48587160)
Also related: Police Have Reportedly Used License Plate Readers to Stalk Romantic Interests at Least 18 Times in Recent Years (https://ij.org/police-have-reportedly-used-license-plate-rea...)
Re:ACLU: Do Automatic License Plate Readers violate privacy under the ...
on Thursday June 18, 2026 @11:33AM (#48586972)
That's a great point regarding other camera networks like Ring. I'm not sure which ones bother me more.. probably Ring but it's close. Seems like we can't escape being surveilled 24/7 everywhere we go. :(
ACLU: Do Automatic License Plate Readers violate privacy under the 4th Amendment
ANEForge: Python for direct computation on the Apple Neural Engine
Algorithmic Information Theory Data Compression Challenge
Re:Cisco Workforce Reductions
on Wednesday May 13, 2026 @11:33PM (#48130801)
The earnings report does mention profit as well. For example, it states: "GAAP EPS of $0.85, up 37% year over year ..."