Samah Abdelrahim سماح

PhD student @StanfordPsych. A developmental psychologist #LangCog lab, interested in everything about children, particularly how they acquire meaning.

[Nassim Nicholas Taleb @nntaleb]

Friends, the problem of statistics is that even if they worked (they don't), whatever is derived statistically ONLY applies to groups never to single individuals.
You are not hiring a subcontinent, but a single individual. Variance determines the max.

[Nuseiba Alarabi @NuseibaAlarabi]

We know exactly how you are trying to undermine the army in order to save the UAE-backed militia. You do this every time the army makes progress and achieves a victory, and we have witnessed this pattern throughout the entire period of the war. Your position toward the militia is deeply suspicious, especially when they have killed, displaced, and raped civilians, and you have the evidence to prove it. But evil will not last forever.

[Katharine Houreld @khoureld]

My piece on the Colombian mercenaries - and their paymasters in the Emirates - is out as the cover story of the Financial Times magazine today. It's the culmination of months of interviews and data tracking with @just1nlynch https://t.co/DRslTYCOws

[Sky News @SkyNews]

Analysis: Sudan war: The UAE's role in this conflict is under increasing scrutiny - and so is Britain's

https://t.co/jlybyd0hAi

[Samah Abdelrahim سماح @SamahARahim]

The bullet that kills you, you never hear." A man went out for water on 8 August 2023 and spent the day sheltering from shells in a stranger's house, drinking tea. At the end, he wanted a handful of ful. Mohamed Nawwar wrote this. I only translated.
https://t.co/u3qRR7fsO2

[Anthropic @AnthropicAI]

While the values Claude expresses shape millions of conversations every day, we don't yet understand why they vary, or whether that's desired.

This approach will allow us to determine what factors influence Claude's value expression—and ultimately how (and whether) to steer it. https://t.co/XGjrkR59Aq

[Robb Willer @RobbWiller]

I really appreciate the rigor and candor of this analysis. This is exactly the posture and approach we need from AI labs (but unfortunately do not usually get).

[Furong Huang @furongh]

While FIFA has captured everyone’s attention, something consequential happened in academia this week, especially for international students in PhD programs.

DHS issued a final rule changing how long F-1 students may remain in the United States.

A thread on what it means and what students and universities can do. 🧵

[Samah Abdelrahim سماح @SamahARahim]

I had just started in a new country, language, and field when I entered his language development class. With 0 background before, I left wanting a lifelong career in it. I will always remember a brilliant scientist, an inspiring teacher, and a kind human.

https://t.co/EZF9im5IHX

[Samah Abdelrahim سماح @SamahARahim]

@clairebergey Congrats Claire 🎉

[Human Rights Watch @hrw]

As evidence of the UAE’s military support to Sudan’s abusive Rapid Support Forces mounts, the NBA will prominently don Emirates logos during the 2026 NBA finals.

The NBA should use this high-profile opportunity to speak out about the country’s human rights record.

Learn more: https://t.co/lXpnoo2J7d

[Marcel Binz @marcel_binz]

Excited to announce that I'll be starting my own lab in Tübingen this October!

Hiring at all levels: Postdoc, PhD & RA.

Want to work on computational cognitive science at scale? Apply: https://t.co/cGX9UfVjwo

Reposts and shares much appreciated 🙏

[Valerio Capraro @ValerioCapraro]

Students without access to LLMs are 2 to 8 times more creative than students with access.

That is the finding of a new paper comparing 2,200 college admissions essays written by humans before ChatGPT with essays generated by GPT-4.

The key point is not individual creativity. GPT-4 can write well, sometimes better than individual students. The problem is collective creativity.

Each new human essay added new semantic territory. New ideas. New angles. New experiences. New combinations.

Each new GPT-4 essay added much less.

The authors call this the diversity growth rate: how much novelty each additional text contributes to the collective pool of ideas.

Humans kept expanding the pool. GPT-4 made the pool converge.

Even when the authors pushed GPT-4 to be more creative, changed parameters, or used chain-of-thought prompting, the homogenizing effect remained.

This is the real danger of AI in education.
Not that students will write worse.
That everyone will write the same.

*
Full paper in the first reply

[Michael Pleyer @symbolicstorage]

In a super cool study Smith et al. (2025) showed that in a communication game, participants who have to build a communication game from the ground up, without the use of language, can make use of salient associations for semantic extension. 5/
https://t.co/fMrStFmQG5

[Marc Owen Jones @marcowenjones]

8/ The report also raises concerns about the future intersection between bot networks and AI systems themselves. As more people rely on LLMs for information discovery, coordinated propaganda ecosystems may increasingly shape the retrieval environments these systems depend on.

[Samah Abdelrahim سماح @SamahARahim]

I just published A leap of faith, and evidence? https://t.co/REccgpzRLa

[Andy Hall @ahall_research]

@timurkuran @profgoose Agreed! Here’s my proposal for how to adapt teaching to AI based on the experimental class I’m teaching at Stanford. Basically: give students coding agents and have them build their own evals.

[Nicholas Kristof @NickKristof]

The world's worst humanitarian crisis in Sudan, and while there are multiple reasons one of the big ones is the role of the United Arab Emirates in supporting a brutal militia. The UAE should be held accountable.

[Eddie Thomas @eddiethomas88]

Good news from Sudan! Prof Awn al-Sharif's magisterial Dictionary of the Colloquial Dialect in Sudan is now an online publication!

[Ev (like in 'evidence', not Eve) Fedorenko 🇺🇦 @ev_fedorenko]

We don't start out with a bilateral language system: it's already strongly left-lateralized in young kids! Resilience of language to early LH damage must occur in spite of this early hemispheric bias. Congrats to Ola @olaozpa and Amanda @Amanda_M_OBrien! Out in NatComms now! 🎉