Frontier AI Research

Update: The external AppView is stable again, so W has switched back. All AT Protocol users are now visible again. Our own AppView worked…

Some W users experienced problems today because W was relying on Bluesky’s AppView. When their AppView went down, parts of W went down with…

W — Trust your feed

Even as @wsocial.eu onboards more and more users, they can’t seem to get them to talk to each other 🤐 This atrocious engagement rate (~9…

W Social is a textbook case for everything wrong with digital sovereignty

We have exciting news: the 🇪🇺 State of the Union speech will be broadcast live on W. And there’s more coming around live video on W. …

The Public Accountability Initiative –LittleSis –

The Uncensored Library – Reporters without borders

Journal publishes paper by video-gamers who derived scientific rules from Stanford online RNA game

Frameshifter

Practical recommendations from a multi-perspective needs and challenges assessment of citizen science games

Arvind Narayanan on Twitter / X

Arvind Narayanan on Twitter / X

Large language models are not the problem

AI Is Not Conscious, But It Is Our Unconscious

AI Is Not Conscious, But It Is Our Unconscious

Technology is a Siren Song

Where Should Science Go Next

Dr. Casey Fiesler (@cfiesler.bsky.social)

How AI assistance impacts the formation of coding skills

First impressions: Delta - Cameron's writing

Introducing Delta

On genAI: Was prototyping really a bottleneck?

My impression is that AI coding is on a "pick two of three" triangle: scope, ship speed, and correctness. Small tools are (speed + correctne…

Devtools must be open source - exe.dev blog

Of Swarms and Sand Gods

Inducing language models to assert their own consciousness restores human beliefs and values

See what you think

Reason Commons | Reason Commons — Issue Trees & Logical Thinking Process

What's the point of a note-taking app?

Note-Taking and Personal Knowledge Management — Unattributed

Logial Thinking Process (LTP) / Issue Tree App and Exploration - Issue Trees & Logical Thinking Process

Own the source. Compose with anything. - The Way of Markdown

Citing Less Critically: LLMs Reshape the Rhetoric and Reach of Scientific Citation

Arvind Narayanan on Twitter / X

AI Research Evaluation: Negative Findings and Failure Modes | Arvind Narayanan posted on the topic | LinkedIn

Leiden Declaration on Artificial Intelligence and Mathematics

Can AI agents conduct open-ended AI research? Early evidence from two case studies

What can we learn from automating an entire quantitative social science paper, from prompt to finished product? Thread about ongoing work…

"... our desire and attention falls to certain things because of the way that we feel about the person, place, or thing that introduces it…

Capturing our Attention

www.sebastianwatzl.com

It's getting harder to focus every day

Almost anything you give sustained attention to will begin to loop on itself and bloom

Movements of the Mind: A Theory of Attention, Intention and Action

Alex Imas on Twitter / X

What Humanity Needs To Flourish In The Next Decade

Pax Machina — Institutions in a World with Powerful AI

Science: A New Golden Age

How AI Agents are transforming scientific discovery

We May Be Entering A Second Axial Age

riding the internet subway to recent links people have visited..

conrad house on Twitter / X

Creative Reading: Scaffolding Reading for Transformation

Wikipedia (@wikipedia.org)

Semble: Rediscovering the Magic of Trails - ATmosphereConf 2026

TrailMarks

The new way we’ll do science

Why we’re expanding the openRxiv Labs Curvenote Reader experiment

Curvenote Reader brings a better reading experience to half a million openRxiv preprints

Launching openRxiv Labs - openRxiv

Universal Scientific Protocols, Inc.

The Micro-Paper: Towards cheaper, citable research ideas and conversations

Genspace

Mamdani Issues AI Moratorium In NYC Schools

Jeff Melnick on Twitter / X

Trying out the features in Claude for education. Join me as I see how it "helps" me write a research paper. My chosen topic: "AI and educati…

Generative AI without guardrails can harm learning: Evidence from high school mathematics

I knew my writing students were using AI. Their confessions led to a powerful teaching moment | Micah Nathan

Luiza Jarovsky, PhD on Twitter / X

Do Drunk Mice Sire Weird-Looking Offspring? – Quentin André

Curious about the competing interests statesments for @i4replication.bsky.social’s @pnas.org paper. 1) funding for i4r comes heavily from an…

Dan Singleton 🏳️‍🌈🏳️‍⚧️ on Twitter / X

Is the map of global bat distribution in this paper (Nov 2025, ‘Sustainability’ journal) AI nonsense? Comments from experts on LinkedIn are…

Trans Journalists Association Stylebook and Coverage Guide

AI Use Disclosure resources