Paul Allen 🌐🇨🇦

Murdering random Jews in response to Israeli state violence is violent antisemitism. It shouldn't be necessary to state that. And yet

Sage Journals: Discover world-class research Subscription and open access journals from Sage, the world's leading independent academic publisher.

"Authoritarian Persuasion at Home and Abroad: The Partial Effectiveness of Foreign Influencers in Propaganda Work"

by Siyu Liang & @lachlanmcnamee.bsky.social in @cpsjournal.bsky.social (2026)

#SpringReading

journals.sagepub.com/doi/10.1177/...

There is a race to the bottom going in with Freedom of Information laws across the country, at every level of government.

A couple of years ago, British Columbia tripled processing fees, justifying it by saying that Alberta charged the same fee.

SparseNUTS: Preconditioning hierarchical models in HMC with a sparse “Laplace approximation” at the marginal mode
statmodeling.stat.columbia.edu/2026/03/12/s...

Silence is not golden when it comes to science and AI The scope of scientific inquiry is contracting as research focuses on data-rich problems

‘Governments investing in AI should consider making equivalent investments in the data collection that AI currently cannot reach.’

In my world that includes means paying more attention to libraries and archives
www.ft.com/content/0c63...

Is Attention Explanation? An Introduction to the Debate Adrien Bibal, Rémi Cardon, David Alfter, Rodrigo Wilkens, Xiaoou Wang, Thomas François, Patrick Watrin. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume ...

Or you can jump straight into the full history paper!

Attention is not Explanation Sarthak Jain, Byron C. Wallace. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short...

Another example is Attention is not Explanation, followed by Attention is not not Explanation, and then all the authors collaborated on a third paper, Learning to Faithfully Rationalize by Construction
aclanthology.org/N19-1357/
aclanthology.org/D19-1002/
aclanthology.org/2020.acl-mai...

Very happy to see this come out 😃

I was just following ... Executive Orders.

These doge men are idiots. They should be held accountable for their actions of ignorance and hubris.

It’s very strange that many headlines today are a about war making petroleum more expensive rather than war killing people, harming people, destroying the environment.

In short: People, not platforms.

The Logic of Connective Faction: How Digitally Networked Elites and Hyper-Partisan Media Radicalize Politics Across democratic systems, ideological cleavages increasingly emerge not only between but also within political parties. At the same time, hyperpartisan and digitally networked media ecosystems amp...

Afterwards @curdknupfer.bsky.social gives a convincing argument on why institutions need to find better ways to understand information flows (you can read more about his work here: www.tandfonline.com/doi/full/10....)

The complexity of misinformation extends beyond virus and warfare analogies - npj Complexity npj Complexity - The complexity of misinformation extends beyond virus and warfare analogies

In the second panel of today, first @andre-k-rodarte.bsky.social and I presented first insights from a recent project studying (1) the susceptibility to misinformation across the complex system through which misinformation flows (www.nature.com/articles/s44...)

I Watched 6 Hours of DOGE Bro Testimony. Here's What They Had to Say For Themselves The hours of videos provide fascinating, or perhaps horrifying, insight into the thinking of someone inside DOGE.

I have watched 6 hours of a DOGE bro deposition. The whole thing, start to finish. It's a fascinating, and horrifying, look at the people who caused so much damage, didn't really achieve anything at all, while still being so sure they are right. Here's what they said: www.404media.co/i-watched-6-...

Biased AI writing assistants shift users’ attitudes on societal issues Biased AI writing assistants shift people’s attitudes about societal issues; common interventions do not prevent this influence.

Ugh. Study shows use of AI can nudge our beliefs...

Are we moving toward a corporate controlled and (given who owns AI) an increasingly conservative hive mind?

Study: Biased AI writing assistants shift users’ attitudes on societal issues www.science.org/doi/10.1126/...

This declaration from Paul Smith, Anthropic’s CCO, gives the flavor. The dispute is over autonomous lethal weapons & mass surveillance of US citizens. Yet the govt is sabotaging Anthropic's contracts with grocery chains, pharma, financial services, fintech ...
2/3

The Women Leaving the New Right Defectors say the movement has dropped the pretense of protecting women and is now openly “cruel and fickle.”

One thing that can tell you quite a lot about where the movement is going is that these stories sound an awful lot like the ones you'd hear from women leaving the white nationalist movement in 2018 on.
nymag.com/intelligence...

Looking to take your data visualization to the next level? The ggdist package extends ggplot2 with specialized tools for visualizing distributions, uncertainties, and comparing data groups.

Visualization: mjskay.github.io/ggdist/

statisticsglobe.com/online-cours...

#tidyverse#package#bigdata

Deconstruction of a memory engram reveals distinct ensembles recruited at learning - Nature Neuroscience Pouget et al. identified distinct CA1 neuron ensembles active during specific moments of fear learning and uncovered the core engram essential for memory formation.

How does the brain build a memory?
A common assumption is that the neurons activated during an experience collectively form the memory engram.
In our new Nature Neuroscience paper (finally out!), we show that this is not the case.
www.nature.com/articles/s41...

How do development and genetics shape evidence integration? 🧠 In our new preprint, we dissect how sensory evidence accumulation over time guides decisions.

Want to know more? Keep reading or or come to Roberto's poster 1-003 this Thursday at #Cosyne2026. (1/4)

Preprint: doi.org/10.64898/202...

How the National Security Strategy Gets Made “Leverage is the favorite verb of the Washington policy community.”

[Classic] "When I first got to the White House in 2017, I didn’t get a “how-to” of how you create a strategy. You read different assessments and past strategies, and you develop an idea of how you might construct one" www.statecraft.pub/p/how-the-na...

Red Flag Alert - Anti-Trans Genocide in the USA - #3 | Lemkin Institute

The Lemkin Institute has issued a *THIRD* trans genocide warning, noting the next stage is widespread containment - and this was before the 4th Circuit ruled today that states can essentially do whatever they want to make trans people "appreciate their sex".
www.lemkininstitute.com/red-flag-ale...

Anthropic announces think tank to examine AI's effect on economy and society The company is doubling down on its ethics stance with the new Anthropic Institute.

According to Anthropic, the institute “will engage with workers and industries facing displacement, and with the people and communities who feel the future bearing down on them but are unsure how to respond.”

Whistleblower claims ex-DOGE member says he took Social Security data to new job The Social Security inspector general’s office is investigating allegations that the former DOGE engineer took sensitive data on a thumb drive in a major potential security breach, said people familia...

"He told a colleague, who refused to help him upload the data because of legal concerns, that he expected to receive a presidential pardon if his actions were deemed to be illegal..."

www.washingtonpost.com/politics/202...

YouTube needs to give this new AI deepfake-detection tool to every user, not just those who can regulate it.

Moms Are Risking Arrest to Protect Gender-Affirming Care for Trans Youth The government wants to ban care nationwide, and hospitals are shutting down treatment. Parents just want it all to stop.

www.them.us/story/trans-...

Good words of warning here. AI models are generally designed to read and incorporate (and, in a sense, believe) virtually everything they find online, with very few guardrails. The information can get divorced from its original source, context, & rebuttals or retractions. That's a really big problem

Modeling Language with Plaster 💡Nerd Rating 3.75/5: This post is about a late-19th-century debate in academic mathematics, but it's plainly written and relevant to LLMs. What is a model, anyway? Doing math used to involve touch....

LLMs “seem to be systems that take the prompt, abstract it into longer and longer texts, connecting to wider clusters of text, and then restructuring all of that text into condensed variations structured like a response.”
mail.cyberneticforests.com/modeling-lan...

A country facing multiple credible threats to its sovereignty but that also cares this little about its own history isn't long for this world.