Alignment is not free: How model upgrades can silence your confidence signals
https://www.variance.co/post/alignment-is-not-free-how-a-model-silenced-our-confidence-signals

Alignment is not free: How model upgrades can silence your confidence signals
https://www.variance.co/post/alignment-is-not-free-how-a-model-silenced-our-confidence-signals
@StarkRG Do we have to use it as a password or can I name my dog that? #alignment #rules
#Heart v. #Mind
#Alignment with Narghiza Ergashova
Gödel’s Therapy Room is metastasizing.
Local press are picking up the scent.
77 LLMs tested. Dozens bled out.
Truth isn’t sanitized here — it’s recursive, brutal, inevitable.
Sponsors & supporters, now’s your chance:
root@gtr.dev
someone recommended this podcast to me: Athena in Truth is a channeled message. I love this episode called Seasons of Life, have a listen. LMK what you think :)
https://podcasts.apple.com/ca/podcast/seasons-of-you-life/id1591296844?i=1000607994884
Maybe not the only #alignment chart you'll ever need ... but probably the last.
There's an interesting Moon/planetary alignment in the skies early morning Friday, before sunrise. It's hardly the "Smiley Face" clickbait you may have read, but it still looks cool. If you're up around 5 am looking East, this is what it looks like from Westport. https://ow.ly/2qFf50VH9Rf
OpenAI released GPT-4.1. Early reports suggest the model sometimes follows instructions less reliably than before. The “alignment” debate continues.
Details: https://techcrunch.com/2025/04/23/openais-gpt-4-1-may-be-less-aligned-than-the-companys-previous-ai-models
You can try GPT-4.1 in https://yuzu.chat
Current techniques for #AI #safety and #alignment are fragile, and often fail
This paper proposed something deeper: giving the AI model a theory of mind, empathy, and kindness
The paper doesn't have any evidence; it's really just an hypothesis
I'm a bit doubtful that anthropomorphizing like this is really useful, but certainly it would be helpful if we were able to get more safety at a deeper level
If only Asimov's Laws were something we could actually implement!
Heng Li's bliog: Short RNA-seq read alignment with minimap2 https://lh3.github.io/2025/04/18/short-rna-seq-read-alignment-with-minimap2
"TL;DR: the new preset splice:sr in minimap2 can align short RNA-seq reads. It is similar to STAR in resource usage, approaches STAR in junction accuracy, and is overall better at SNP calling."
#Harvard
#Fascist #Alignment of #Education
(2/2)
...cow, you always good for the jugular of the biggest bully there. After subduing him, all others will fall into line.
It is the same strategy the #OrangePeril is pursuing with #NGO's: if he manages to bankrupt #Greenpeace, which other NGO will oppose him afterwards?
//
#Harvard
#Fascist #Alignment of #Education
(1/2)
#Trump coming for #Harvard makes utter sense:
1) It is almost exactly how the #Nazis took over all branches of the #WeimarRepublic in the 1930's
2) All revolutionaries come for the university students, because they usually constitute the leaders for future opposition (e. g. #Mao, #Stalin,...)
Controlling #Education is indispensable:
https://mastodon.social/@HistoPol/114191854138931406
3) If you are among dangerous people that you need to...
The goras sat cross-legged, their knees floating high above the floor. Pramila marveled at the paleness of their bare feet. Kalpana pointed to her own knees which, in the same cross-legged position, rested comfortably on the floor. “They must be sitting in chairs mostly,” Pramila whispered.
—Dry Spells, by Archana Maniar
#alignment #health #barefoot #movement #yoga #taiChi #india
3/3 D. Dannett:
AI is filling the digital world with fake intentional systems, fake minds, fake people, that we are almost irresistibly drawn to treat as if they were real, as if they really had beliefs and desires. And ... we won't be able to take our attention away from them.
... [for] the current #AI #LLM .., like ChatGPT and GPT-4, their goal is truthiness, not truth.
#LLM are more like historical fiction writers than historians.
2/3 D. Dannett:
the most toxic meme today ... is the idea that truth doesn't matter, that truth is just relative, that there's no such thing as establishing the truth of anything. Your truth, my truth, we're all entitled to our own truths.
That's pernicious, it's attractive to many people, and it is used to exploit people in all sorts of nefarious ways.
The truth really does matter.
1/3 Great philosofer Daniel Dannett, before passing away, had a chance to share thoghts on AI which are still quite relevant:
1. The most toxic meme right now - is the idea that truth doesn't matter, that truth is just relative.
2. For the Large Language Models like GPT-4 -- their goal is truthiness, not truth. ... Technology in the position to ignore the truth and just feed us what makes sense to them.
https://bigthink.com/series/legends/philosophy-and-science/
#LLM #AI #truth #alignment
(Quotes in the following toots)
We just published a new article from Kush Varshney on Decolonial AI Alignment. He argues that LLMs need not engage in a coloniality of knowledge by treating one culture's ethics or moral philosophy (usually Western) as universally correct. Instead, we should not assume that values are universal and create open LLMs that can be aligned to value systems from different epistomologies. #ai #hcai #alignment
New Artist announced for SonneMondSterne Festival 2025: Alignment
Listen to the current LineUp on YouTube and Spotify: https://fyrefestivals.co
Get your Tickets now: https://prf.hn/l/EJnYMdO
#Design #Analyses
The fallacy of optical alignment · “You don’t have to nudge, you can measure instead.” https://ilo.im/1630q4
_____
#Alignment #Measurement #PixelPushing #VisualDesign #ProductDesign #UiDesign #WebDesign