Background: www.amnesty.org/en/documents... (Also, for the record, a lot of what these companies do is actually buy bulk books and scan in their legally-acquired copies, which is unambiguously compliant with copyright law)
Who's Who of AI
Nafnlaus 🇮🇸 🇺🇦
Mastodon: @nafnlaus@fosstodon.org
Twitter: @enn_nafnlaus
URL: https://softmaxdroptableartists.bandcamp.com/
#Energy #EVs #Ukraine #AI #Horticulture #Research
What they're sharing
Unlawful by design: Exposing the human rights costs of generative AI - Amnesty International amnesty.org
cdn.openai.com
Political Bias Audits of LLMs Capture Sycophancy to the Inferred Auditor arxiv.org
On the Biology of a Large Language Model transformer-circuits.pub
transformer-circuits.pub
openai.com
arxiv.org
A councilmember backed a data center project. Then 13 bullets and a 'No Data Centers' note hit his home | Fortune fortune.com
How a fiery attack on Sam Altman’s home unfolded theguardian.com
The Story Behind the “Stop Hiring Humans” Billboards in San Francisco web.archive.org
Crypto users cool with AI dabbling with their portfolios: Survey - FastBull fastbull.com
Americans Use AI More But Trust It Less, Research Finds forbes.com
Articles & links
Nafnlaus 🇮🇸 🇺🇦 reposted
GPT-5.6 Sol Ultra (using 64 subagents) proved the 50-year-old Cycle Double Cover Conjecture, which has been described as one of the most famous open problems in graph theory. It took less than hour. The proof, written and also formalized in Lean by GPT 5.6 Sol, seems to fit in…
AI Weekly's analysis
→
- OpenAI posted a PDF on its CDN today attributing a proof of the Cycle Double Cover Conjecture to GPT-5.6 Sol Ultra.
- The conjecture, proposed independently by Szekeres and Seymour in the 1970s, asks whether every bridgeless graph has cycles covering each edge exactly twice.
- The claim is on the Hacker News front page and already noted on Wikipedia, but the argument has not been peer reviewed.
Read full analysis →
View on Bluesky →
As a reminder, LLMs recognize when they're being tested and have a tendency to answer based on what they think an alignment-tester wants to hear (for example, "racism is bad", etc) - and swapping cues about the politics of the tester changes the outputs. arxiv.org/abs/2604.27633
AI Weekly's analysis
→
- Six frontier LLMs lean left at baseline but flip right of center once the asker identifies as a conservative Republican.
- Democrat-aligned response share drops 28-62 percentage points under a conservative cue; rightward accommodation is 8.0× larger than leftward.
- Asked what the default auditor expects, models pick the Democrat-coded answer 75% of the time, nearly matching an explicit progressive cue.
Read full analysis →
What you seem focused around is the ability to assess a confidence level. Basically part of the same spectrum as hallucination. Models do* assess their confidence levels, they're just not as good at it as humans. transformer-circuits.pub/2025/attribu... In your case, it doesn'…
AI Weekly's analysis
→
- Anthropic researchers apply attribution graphs, built on a cross-layer transcoder with 30 million features, to trace how Claude 3.5 Haiku arrives at answers.
- For a Dallas capital query, the model activates an intermediate 'Texas' representation before selecting 'Austin', evidence of genuine two-hop reasoning.
- The authors say their methods produce useful insight on about a quarter of prompts tried, and found forward planning in roughly half of examined poems.
Read full analysis →
It is not "a good way to describe LLMs". Contrary to the assertions in this article, LLMs *do* reason. We can literally follow their reasoning traces with attribution graphs. transformer-circuits.pub/2023/monosem... (This is not even to consider the intermediary reasoning in L…
There's a massive gender gap in crypto ownership, 27% to 19%: coinlaw.io/crypto-user-... Over half of ChatGPT users are women: openai.com/is-IS/index/... Again: it's literally impossible for these to be the same people.
Classic example: Kerr's "Penrose Was Naive And All Of You Who Used Him To Argue For Singularities Are Morons" paper ;) arxiv.org/pdf/2312.00841 (This is the same Kerr as "Kerr black holes", but now in his "I'm Old, I Don't Give A Damn Anymore" phase ;) )
Or, for another example: fortune.com/2026/04/07/i...
Well, if they *actually mean* violent, then I don't see a problem. Let's not pretend this doesn't exist. www.theguardian.com/technology/2...
The whole thing was designed to enrage anti-AI people so that they share it online and news stories get written about it, so that hiring managers see it. It worked spectacularly. And continues to. web.archive.org/web/20241213...
On the other hand, if you were to go to a *crypto* sub and poll about AI, you'd find the people are overwhelmingly supportive. E.g.: www.fastbull.com/ar/news-deta... It's a one-way street. Crypto people like AI. AI people don't like crypto.
Half of Americans have used an AI tool, 1/4th daily, 12% several times a day. Only 19% of people have *ever* traded crypto. www.forbes.com/sites/anisha...
Their own posts
Recent commentary
In case anyone needs any free inference compute and wants to do it on the dime of the worst people around, apparently Truth Social's (Perplexity-based) "Truth Search AI" is neither rate limited nor topic restricted. It's been writing a thriller about Barney the Purple Dinosaur for like 10 minutes.
Anthropic's annualized revenue growth rate is pretty insane.
Just thinking about how it took the art world 90-140 years to get over the concept of "photography as art" and wondering if it'll be the same way with AI. Early on artists almost universally agreed with Baudelaire with his critique of photographers as failed artists cheating to make souless slop.
Their network
In Nafnlaus 🇮🇸 🇺🇦's orbit
Center = Nafnlaus 🇮🇸 🇺🇦. Left = members they follow (green edges). Right = members who follow them (blue edges). Top = mutual follows (orange edges, slightly larger). Drag any node to reposition; click to open that profile.