Ordinary social networks face a constant onslaught of chatbots pretending to be human. A new social platform for AI agents, called Moltbook and designed to look a lot like Reddit, may face the opposite problem: getting clogged up by humans pretending to post as bots.
www.theverge.com/ai-artificia...
Epstein files as Bataille’s The Accursed Share in action: bacchanalia of elite debauchery, financed by superprofits from post-911 military Keynesianism.
Missions succeed when public finance delivers public value.
While the UK is on track to deliver its Clean Power mission w its expansion in offshore wind, it must learn from the failures of PFI era and financialisation of utilities like Thames Water: don’t socialise risks & privatise rewards.
Some quick footnotes:
* A summary of the majority view from insiders in the tech industry on what's really going on with AI www.anildash.com/2025/10/17/t...
* A reflection on what a "good" AI system would have to look like www.anildash.com/2025/05/01/w...
I think the answer to stopping the harms of the Big AI companies is *not* to say "we need to kill all the AI", even though we hate seeing the creeping impacts of that bad AI in our lives. The model has to be how Bluesky is a better alternative to Twitter. We need a tolerable AI alternative instead.
Because hundreds of millions of people are already using these tools and finding utility in them. And they're not crazy, they are reasonable. Telling them, "no, you have to do without", will make them think, "Oh, I guess the critics are crazy and the AI bros are the reasonable ones."
This year has been brutal for science, especially early career researchers and those working with communities further marginalized by this administration. Please read the experiences of these scientists.
And when you’re done reading, find a way to fight back.
theconversation.com/this-year-ne...
The planned hepatitis B birth-dose trial in Guinea-Bissau raises serious ethical concerns. Withholding a proven, life-saving vaccine from newborns to answer speculative questions is an absence of equipoise with real downstream harms for trust in vaccines.
bktitanji.substack.com/p/how-unethi...
We're advertising a PhD position! 😎 Modelling the epidemiological impact of HIV prophylaxis on a backdrop of social, cultural, and behavioural complexity
www.findaphd.com/phds/project...
It's more like, conglomerates establish a bank, & once they do, odds are stacked in their favor. Every monetary injection via RRR cut (and maybe even every withdrawal via repo rate hike, though I'm yet to explore this), conglomerates benefit - over & above MSMEs & workers. BSP's fiat is their mana.
“… as Congress failed to raise the minimum wage by a couple of hundred pesos & facilitate a transfer of a measly ₱259 billion to close the ₱5 trillion wage-profit gap, the BSP has effectively subsidized capitalists to the tune of ₱300-400 billion.”
medium.com/@futilityfun...
Fitting a generalized mixed model with a gamma distribution log link and random slopes to reaction time data to arrive at precisely the same point estimate as the authors did by simply averaging and conducting a t-test:
CDC ends COVID vaccine recommendation for healthy kids and the pregnant. HHS cancelling Moderna contracts for developing vaccines vs. future flu pandemics. RFK, Jr. may ban US scientists from publishing. #EpiBluesky
Workers don’t only hold jobs; they create them - every time they buy, save, discover, or create. So-called “financial strategy” is mere exploitation of signals by workers to workers.
Stancil and the leftists have been shouting the two parts of this sentence at each other for years now when they should just be shouting the name at the top to everyone else
🆕 New week, new pub! @sarahlotspeich.bsky.social (Assistant Prof) published in Statistics in Medicine with @wakeforest.bsky.social colleagues @ashleymullan.bsky.social (MS alum, now PhDing @vanderbilt.edu), @lucystats.bsky.social (Assistant Prof), and Staci Hepler (Vice Chair/Associate Prof).
Flyer for International Social Epidemiology Webinar Series: Professor Naomi Priest from the Australian National University on 1st of May. Title: "My 'accidental' career as a social epidemiologist"
📢 Mark your calendars for our next webinar next week!
“My 'Accidental' Career as a Social Epidemiologist” featuring Professor Naomi Priest
📅 Thursday, May 1st, 2025
🕗 06:00 New York / 11:00 London / 20:00 Melbourne
👉 Register now at bit.ly/3YABrxB — and please share widely!
#SocialEpi #SDoH
that feel when you can’t reconcile the reality of historical contingency with the possibility of inductive inference
This is type of cool and quiet science that excites me. In 2023, #H5N1 hit California condors hard: >21 birds dead. Just ~560 remain globally. Researchers asked: could a poultry vaccine help? They ran a trial that might change wildlife conservation & save the species.🧪
wwwnc.cdc.gov/eid/article/...
Peruvian Population Genomics: Unraveling the Genetic Landscape and Admixture Dynamics of Urban Populations www.biorxiv.org/content/10.1101/2025.04....
My recommendations on economic growth and development
P.P.S. How much less? 100.08%/100.00016102385% = 0.99206508952. So the probability of being selected shrinks by 0.21% a year. By the end of 60 years your probability of being selected - little as it is - would have shrunk to 62% of the original (P = 0.00002173913). 11/10
P.S. Note that is an understatement. That the population growth rate in the PH is 0.8% a year, but the required sample size for a 2% margin will only grow by 0.00016102385% a year! That means as the pop'ln grows you have less & less chances of being selected per year. 10/10
So, if you get interviewed at least once in your entire electoral life, consider yourself the lucky one out of five Filipinos. Otherwise, that's just the consequence of statistics. 9/
So our formula gives us 1-(1-P)^9,360 = 0.18411386371.
That's roughly 18.5% - there is only less than 1 in 5 chances that you will be interviewed at least once for the rest of your electoral life, even if we assume there are 3 surveys a week. 8/
Do the same to compute the prob. of one being interviewed at least once in his entire life in all major surveys. Say there are at least 3 surveys per week from different firms. That's extreme. Say that one's electorally active life is 60 years. So that's 3x52x60=9,360 surveys. 7/
From here, we can get the probability that you were interviewed for at least one of the surveys. That is 1-(1-P)^192 = 0.00416525948, or 0.416525948% - less than half of a percent chance. 6/
