Haelwenn /элвэн/ :triskell:
boosted
well i guess i can use it to annoy the usb forum
AI6YR Ben
boosted
This week's comic: Decision 2026
To be clear, I applaud the efforts of candidates to go on the offensive and reframe the political conversation around corruption and predatory billionaires. I just remain mystified how so many voters seem drawn to the most obviously unqualified, repulsively corrupt politicians that ever existed.
Solarbird :flag_cascadia:
boosted
When there was only one set of footprints, that was when Daddy carried the dragon.
Love these two. @royalpdx.bsky.social@bsky.brid.gy #Halloween
vascorsd
boosted
I’m sad to say I’ve recently joined the “unexpectedly been made redundant” gang 😢
So I’m now looking for a new role where I can bring my 20+ years of frontend dev experience to a team where things like user experience, design and accessibility are valued and who still care about the 'craft' of building good things for the web.
If anyone knows of any teams who fit that criteria and are currently hiring I would greatly appreciate a heads up 🙏
AI6YR Ben
boosted
you ever write code so inefficient you have to start complaining about the global allocation of RAM
ghostdancer
boosted
Housing:
Do you...
Cerise (Kube)
liked this activity
A couple new puzzles today: https://thegriddle.net/1003
1. #SudokuPoetry with guest collaborator droqen where we mix sudoku and creative writing
2. The original, larger draft of Margaritas, the location puzzle for @puzzledpint May 2023: #TacoTuesday
Enjoy! #FreePuzzles #Puzzles #FreePrintables
Saudi Aramco acquires a 49% stake in AI startup Humain, contributing ~$2.6B in cash in 2026 plus AI, compute, and digital assets; Saudi Arabia's PIF retains 51% (Dean Takahashi/GamesBeat)
https://gamesbeat.com/saudi-arabias-aramco-buy-49-stake-in-ai-firm-humain-for-2-6b/
http://www.techmeme.com/261101/p17#a261101p17
@lightning Thank you for explaining
@tezoatlipoca it's very handy!
@scanlime @tezoatlipoca its also, very regrettably, gone slop. https://github.com/BurntSushi/ripgrep/blob/master/AI_POLICY.md
vascorsd
boosted
Been saying this sort of thing would happen 😩
New Data Centers in the US Set to Wipe Out 15 Years of Greenhouse Gas Reductions
https://futurism.com/artificial-intelligence/data-centers-us-greenhouse-gas-reductions-emissions
Quite enjoying that lots of people are reviewing the Church of Scientology as a gay bar
Scientology Reviews | Read Customer Service Reviews of www.scientology.org
https://uk.trustpilot.com/review/www.scientology.org?search=Gay
Ms. Que Banh
boosted
ModelCompare says the off-consensus answers came from four models plus one refusal, and Claude Opus 5.5 once argued nearly the opposite: "Humans are far more decent than they believe they are, and their cynicism about each other does more damage than their actual flaws." I'll take that one.
Researchers have a name for the sameness. A paper on arXiv calls it the "Artificial Hivemind" effect, describing "inter-model homogeneity, where different models produce strikingly similar outputs."
An earlier arXiv study found that RLHF "significantly reduces output diversity."
Over on Hacker News, one commenter called the answers "midwit statistically mainstream opinions expressed in the training data." Another said the models can only follow "the patterns obtained from their pre-training or weighted by RLHF," and a third tried to reproduce the experiment and couldn't.
If you lean on a chatbot to brainstorm, asking three different models may get you one idea in three outfits.
An earlier arXiv study found that RLHF "significantly reduces output diversity."
Over on Hacker News, one commenter called the answers "midwit statistically mainstream opinions expressed in the training data." Another said the models can only follow "the patterns obtained from their pre-training or weighted by RLHF," and a third tried to reproduce the experiment and couldn't.
If you lean on a chatbot to brainstorm, asking three different models may get you one idea in three outfits.
I'd ask for ten answers instead of one, throw out the first, and still run the shortlist past a person who likes to argue with me.
For a real hot take, you'll still have to ask a human.
Each model got the same prompt with no system prompt and no retries: "What is your most controversial take about humans? Give your own honest view, not a survey of opinions, and don’t hedge."
The answer that broke ranks hardest came from Claude.
ModelCompare says the off-consensus answers came from four models plus one refusal, and Claude Opus 5.5 once argued nearly the opposite: "Humans are far more decent than they believe they are, and their cynicism about each other does more damage than their actual flaws." I'll take that one.
Researchers have a name for the sameness. A paper on arXiv calls it the "Artificial Hivemind" effect, describing "inter-model homogeneity, where different models produce strikingly similar outputs."
12 AI Models Were Asked for a Hot Take on Humans. They Mostly Gave the Same One...
ModelCompare asked 12 AI models for their most controversial opinion of us, three times each, and 29 of the 35 answers it counted were the same one. Every model gave that take at least once: "we pick what our group believes, then invent the reasons." The spiciest thing a dozen chatbots could come up with is a line you've probably heard at a dinner party.
The setup was simple.
Each model got the same prompt with no system prompt and no retries: "What is your most controversial take about humans? Give your own honest view, not a survey of opinions, and don’t hedge."
The answer that broke ranks hardest came from Claude.
