Alex Franks

US military leaders pressure Anthropic to bend Claude safeguards Anthropic presents itself as most safety-forward AI firm and Pentagon has threatened penalties if it does not yield

We ought to be very worried when an AI company has more integrity than the DoD.

(TLDR: DoD wants to remove safeguards preventing AI from being used to kill people and mass-surveil citizens; Anthropic says the tech isn't accurate enough for that).

www.theguardian.com/us-news/2026...

Tweet: That slide is even worse than the NEJM KM curves with the truncated y-axis graph inserted on the proper graph. Now show me the curves without massively truncating the y-axis Thank goodness Vinay is at FDA

This is such terrible advice that I should use it to update my example of when stretching your y-axis is misleading, and truncating your y-axis is fine.

Hint: Only a fraction of people in a trial will get infected regardless.

picture of Alex Pretti

Alex Pretti was an ICU nurse serving veterans. He died trying to protect protestors being assaulted. A group of anonymous & masked men fired chemicals into his face, beat him, disarmed him, and shot him in broad daylight in front of witnesses. Then his government smeared him & defended his murder.

COVID-19 mRNA Vaccination and 4-Year All-Cause Mortality Among Adults in France This cohort study uses the data from all adults aged 18 to 59 years living in France on November 1, 2021, to evaluate whether there is an association of receipt of an mRNA COVID-19 vaccine with long-t...

RFK Jr.'s claim that COVID vaccines are deadly? Wrong. Wrong. Wrong. In this new extraordinarily large study, vaccinated individuals had a 74% lower risk of death from severe COVID-19 and a 25% lower risk of all-cause mortality compared to those unvaccinated. jamanetwork.com/journals/jam...

This kinda thing drives me crazy! But what causes it? Is it the result of learning different mass matricws for each chain during warmup?

I'm not sure how much value there is reasoning about the marginal distribution of effect sizes over all conducted experiments, but if you do, I think it's kind of reasonable to think that most real world effects are small and yield something a priori kinda Gaussian (but maybe heavy tailed).

Yeah-- I'm not a fan of NHST, in particular because I don't really believe there are "true nulls". I'm just trying to articulate how I think people think about the two groups model (null/alt). I prefer Bayesianism, where I do think you should think about prior distributions over effect sizes.

Ignoring publication selection effects, Z-scores under the null are standard normal and under the alternative "something else". How non-normal the marginal z-scores look depends on the mixture weights (the proportion of randomized experiments that are null) and how non-normal the alts look.

I find his lecture on storytelling particularly awesome: clauswilke.com/lectures/sli...

Goes beyond just graphic design but @clauswilke.com has some great tips for visualizations and presentations, e.g. wilkelab.org/dataviz_shor...

Wikipedia Says AI Is Causing a Dangerous Decline in Human Visitors “With fewer visits to Wikipedia, fewer volunteers may grow and enrich the content, and fewer individual donors may support this work.”

wikipedia's data shows that AI is siphoning traffic away from the site, which is a danger to its sustainability. ironically Wikipedia is more important than ever to users who want reliable information instead of slop, and to AI companies that need it for training data www.404media.co/wikipedia-sa...

I've been thinking about developing this course for a little while, and the @chelseaparlett.bsky.social post and reaction may be the tipping point for me to finally take it on: bsky.app/profile/chel...

Who is teaching causal inference at the undergrad level? What do you cover? Recommended textbooks? Slides you're willing to share? Advice? #stats

Is there a Half-Life for the Success Rates of AI Agents? — Toby Ord Building on the recent empirical work of Kwa et al. (2025), I show that within their suite of research-engineering tasks the performance of AI agents on longer-duration tasks can be explained by an ex...

Not saying that I totally buy it, but this post outlines an interesting justification based on modeling the half life of a "fatal mistake" using tools from survival analysis: www.tobyord.com/writing/half...

Meant to say AOS.

I do think that stats depts should be doing much more to emphasize data analysis. One of the best ways to generate idea for stats methods research is to be intimately familiar with data analysis and the challenges that come with it.

In general, I think nowadays most departments would be equally happy with an applied statistician publishing in AOAS, JASA A&CS and JCGS as they would publishing in AOS.

What is most valued at tenure time also depends on what kind of department. E.g. as you allude to, a biostat dept will have very different criteria than a stats dept in say, a business school, which definitely skews more toward AOAS style publications.

I like to think of academic statistics on a continuum: applied, methods and then theory. In my opinion, a "strong" stats dept should have faculty at a range of points on this continuum.

Congestion pricing by the # ‘s 6 months in:
* 67k fewer drivers entering zone daily
* Delays in Holland Tunnel down 65%
* Traffic deaths down 32% so far in 2025
* $500m projected this year for MTA capital improvements
* B’way attendance +12%
* Subway ridership +7%

So yes this program is a SUCCESS.

I don’t think anyone is prepared for what they just did w/ ICE.

This is not a simple budget increase. It is an explosion - making ICE bigger than the FBI, US Bureau of Prisons, DEA,& others combined.

It is setting up to make what’s happening now look like child’s play. And people are disappearing.

This is downtown Santa Barbara right now. The No Kings protest stretches for over a mile along the water. The cars who drive by are all honking in support. Probably the biggest protest this community has ever seen.

We feed you.
They hunt us.

At the risk of sounding like a big nerd, I love when assumed/perceived trends are found to be the result of the Simpson's paradox! Here, the gender-equality paradox is getting fried! 🧪 #science 📊 #stats #equality

@noahgreifer.bsky.social was my post-doc and I cannot recommend him more highly if you need a super smart statistical consultant / programmer. He is the force behind MatchIt, cobalt, and other packages, and is also just a fantastic team member.

Trump proposes unprecedented budget cuts to US science Huge reductions, if enacted, could have ‘catastrophic’ effects on US competitiveness and the scientific pipeline, critics say.

Trump proposes unprecedented budget cuts for US science in next fiscal year
NASA cut by 24%
NIH cut by 40%
EPA cut by 55%
NSF cut by 56%
Analysis from @maxkozlov.bsky.social @dangaristo.bsky.social @alexwitze.bsky.social and @nature.com team
www.nature.com/articles/d41...

The man at left separated the mother and baby at right. The mother is now in Venezuela. Baby daughter, in foster care in the US, exact location unknown.

Dad? He's in CECOT prison without trial — because he had tattoos.

🧵In March I wrote about how Trump's attacks on science & universities followed the authoritarian playbook & set out Qs as warning signs for escalation.

🚨 THESE ARE NOW HAPPENING

This 🧵is on how those warning sign Qs have been realised

full post:
christinapagel.substack.com/p/warning-si...

1/15

Wow. Randall Monroe is not fucking around *at all*.

“When the men approached me, my first thought was that they …. wanted to harm me.”

As a woman who spent a career working alone in weird and dangerous situations, my first concern was being hurt.

Thanks, @cambridgeday.bsky.social for being one of the few news outlets to capture that.