Saige Rutherford, PhD

In light of this deeply illegal activity at NSF and the ongoing attack on American science, I'm posting the slides from my 2017 talk at the American Physical Society about the Nazi take over of German physics/why German is no longer science's lingua franca 🧪

🧵 1/26
drive.google.com/file/d/1_MCm...

title and abstract from https://arxiv.org/pdf/2507.19960

table 1 from https://arxiv.org/pdf/2507.19960

Boiling here at home in Cyprus but I put the finishing touches a couple of days ago on this preprint: What Does 'Human-Centred AI' Mean? doi.org/10.48550/arX...

Wherein I analyse HCAI & demonstrate through 3 triplets my new tripartite definition of AI (Table 1) that properly centres the human. 1/n

A common gotcha is "but Olivia, proofs don't capture reality" which I find honestly so beautiful because that's THE point. A formal system, a proof, maths, code, 💯 CANNOT solve reality, the frame problem, human cognition — so their gut tells them exactly the answer. Lean into it! That's exactly it.

After having such a great time at #CHI2025 and #FAccT2025, I wanted to share some of my favorite recent papers here!

I'll aim to post new ones throughout the summer and will tag all the authors I can find on Bsky. Please feel welcome to chime in with thoughts / paper recs / etc.!!

🧵⬇️:

Today (w/ @ox.ac.uk @stanford @MIT @LSE) we’re sharing the results of the largest AI persuasion experiments to date: 76k participants, 19  LLMs, 707 political issues.

We examine “levers” of AI persuasion: model scale, post-training, prompting, personalization, & more! 

🧵:

1/11 Excited to share our @Naturestudy led by @leonooi.bsky.social @csabaorban.bsky.social @shaoshiz.bsky.social

AI performance is known to scale with logarithm of sample size (Kaplan 2020), but in many domains, sample size can be # participants or # measurements...

doi.org/10.1038/s415...

Excited to be at the Computational Psychiatry Conference #CPC2025 in Tübingen! I’ll be presenting a poster called Computational Psychiatry meets Algorithmic Fairness. You can read it (along with some of my other philosophical ideas) here: www.beingsaige.com/single-post/...

I defended my PhD! You can watch it on YouTube here: youtu.be/OCfqTtPy1KM?...

If you want to make America healthy again, start with

Universal health & UBI
Masks in healthcare
Clean air & water
Affordable housing & food
Guaranteed education
Paid time off
Affordable child care

When we accept that caring for one another is essential for health & happiness, we all benefit

There may be even more stupid use cases for #genAI, but this one is definitely a top contender

Congratulations @Saige to a wonderful PhD Defense!

GitHub - KempnerInstitute/nicewebrl: NiceWebRL is a Python library for quickly making human subject experiments that leverage machine reinforcement learning environments. NiceWebRL is a Python library for quickly making human subject experiments that leverage machine reinforcement learning environments. - KempnerInstitute/nicewebrl

To help computational cognitive scientist engage with more naturalistic experiments, I've made NiceWebRL.

NiceWebRL is a Python library for designing human subject experiments that leverage machine reinforcement learning environments.

github.com/KempnerInsti...

ADD YOUR NAME: Sign the Open Letter in Support of NIH Staff Join the courageous and committed National Institutes of Health (NIH) public servants by adding your name now.

What does 18,000+ brave signatories look like standing together in massive solidarity with the #NIH heroes of the historic Bethesda Declaration??...

Find out by adding YOUR NAME to our #NIHLove letter here! -

actionnetwork.org/forms/add-na...

#StandUpForScience
#BethesdaDeclaration
#StopTrump

The cover of The Lancet with this quote from our commentary: "If we in academic medicine give in to the campaigns against diversity, equity, inclusion, and accessibility, the consequences will include resegregation and fewer opportunities for marginalised groups."

Holy smokes! Our article is on the cover of The Lancet!!!!!!!!!! Nbd 💅

Science builds our future.

In Europe it has the freedom and support it needs.

To find the next cure and develop the next ground-breaking technology.

Our mission is clear:

We want the brightest minds and boldest innovators to "Choose Europe!" ↓

europa.eu/!fyy9RX

Multi-timescale reinforcement learning in the brain - Nature Individual dopaminergic neurons encode future rewards over distinct temporal horizons.

Our work with Pablo Tano, @hyunggoo-kim.bsky.social Athar Malik, Alexandre Pouget and @naoshigeuchida.bsky.social exploring how dopamine neurons could enable multi-timescale reinforcement learning in the brain is out in @nature.com
www.nature.com/articles/s41...

The MAHA Report: I Read it so You Don't Have to. Part I: In the Beginning

Curious about the MAHA Report on children’s health but don’t have the time or energy or inclination to read all 75 pages?

You’re in luck! I’m reading it for you. Link in replies.

open.substack.com/pub/epiellie...

Changes to the 2026 and 2027 Work Programmes With the launch of the competitions for grants under ERC Work Programme 2026 in July of this year, several changes to the submission of applications and the evaluation of proposals will apply. The mai...

Planning to apply for #research #funding from the ERC?

From the next application rounds, expect changes to the:

• proposal structure
• evaluation process
• extra funding you can request
• eligibility for Starting & Consolidator #Grants (from 2027)

More 👇 europa.eu/!RPHWvv

Cost-cutting tactics The secret bonuses were just one of many maneuvers UnitedHealth devised to track and cut expenses in its nursing home initiative. Internal emails show, for example, that UnitedHealth supervisors gave their teams “budgets” showing how many hospital admissions they had “left” to use up on nursing home patients. The company also monitored nursing homes that had smaller numbers of patients with “do not resuscitate” – or DNR – and “do not intubate” orders in their files. Without such orders, patients are in line for certain life-saving treatments that might lead to costly hospital stays. Two current and three former UnitedHealth nurse practitioners told the Guardian that UnitedHealth managers pressed nurse practitioners to persuade Medicare Advantage members to change their “code status” to DNR even when patients had clearly expressed a desire that all available treatments be used to keep them alive.

UnitedHealth secretly pressured nurses to change patients' records to include DNRs (Do Not Resuscitate) without their consent, so those patients would die instead of being saved with medical interventions that could cost UnitedHealth money.

This is murder.

www.theguardian.com/us-news/2025...

Impact Dashboard - Impact Counter | Impact Counter Visualize the human impact of funding changes for aid and support organizations.

Musk's chainsaw to programs that support impoverished folks has a body count.

Many thanks to my BU colleague
@brookeenichols.bsky.social
for developing the Impact Counter, an incredible tool that illustrates the invisible violence of policy. This is real science.

103 deaths per HOUR= 2472/day

The history and future of resting-state functional magnetic resonance imaging - Nature This Review provides an overview of the history of resting-state functional MRI research, which has helped to reveal the spatiotemporal organization of the brain, and discusses how it can contribute f...

When I first started working with resting state fMRI as a postdoc, there was a lot of skepticism about what we could learn from it. 20 years later, it's hard to imagine where the field of neuroscience would be without it. Here's a summary 🧠 www.nature.com/articles/s41...

House Republicans want to stop states from regulating AI. More than 100 organizations are pushing back | CNN Business More than 100 organizations are raising alarms about a provision in the House’s sweeping tax and spending cuts package that would hamstring the regulation of artificial intelligence systems.

We’re among 100+ organizations calling on House leaders to reject a proposal that would stop states from regulating AI. This still-in-development technology should not be allowed to make unfettered life-or-death decisions about our lives, safety, and freedoms. edition.cnn.com/2025/05/19/t...

Abstract and title page of PDF

Table 1

Table 2

I've felt for a while that a mainstream method, reverse engineering, in cognitive science & AI is incompatible w computationalism‼️ So I wrote "Modern Alchemy: Neurocognitive Reverse Engineering" w the wonderful Natalia S. & @irisvanrooij.bsky.social to elaborate: philsci-archive.pitt.edu/25289/
1/n

The Pulitzer Prizes Explanatory Reporting finalists: Alexia Campbell, April Simpson and Pratheek Rebala of the Center for Public Integrity; Nadia Hamdan of Reveal; and Roy Hurst, contributor, Mother Jones. Azam Ahmed, Christina Goldbaum of The New York Times and Matthieu Aikins, contributing writer. Annie Waldman, Duaa Eldeib, Max Blau and Maya Miller of ProPublica.

Our “America’s Mental Barrier” series, which examines how insurance companies interfere with access to necessary mental health care across the United States, was also named a finalist in the explanatory reporting category: www.propublica.org/series/ameri...

Via @pulitzerprizes.bsky.social

Proud husband alert 🤗🥰😘

Harvard Medical researcher sues NIH for cutting grant to study LGBTQ mental health "This is not just an attack on the LGBTQ community — it is a blow to the integrity of science and the health of every American," a Harvard Medical School associate professor studying the mental…

⛔ NIH ended ALL of my grants mid-project, given LGBTQ health focus 🏳️‍🌈

⚖️ So what am I doing? Suing RFK, Jr.

As one team member shared in this story👇: "The threats to our work don’t just make me fear for my career — they make me fear for my safety.“

#SaveDemocracy #SupportScience #PublicHealth

My Ph.D. thesis is now online (www.publicatie-online.nl/files/9017/4...), the books are printed, and my defense is scheduled for June 17th. What a surreal feeling! #PhDone

Complete and transparent reporting aids critical and assessing risk of bias in #predictiveAI. TRIPOD+AI and PROBAST+AI are key tools to improve #AI research in healthcare

TRIPOD+AI -> tinyurl.com/39pz3rfd
PROBAST+AI -> tinyurl.com/yt8vrvrf

#Digitalhealth #healthcareAI #machinelearning

Charlie-Mop is ready for the long weekend. 🐣🐇🐑