I really appreciate @elizabethwithaz.bsky.social's reflections on how she explored the social impacts of technology with her students by introducing them to the people - the ghost workers - whose labor makes “AI” systems possible. Read this week’s @civicsoftech.bsky.social and find the lesson at:

BBC reporting that Anthropic has a new AI product so powerful - especially good at hacking - that it will not make it generally available but will let a few companies and maybe the US government have it. What could possibly go wrong?

Research: What China’s AI Agents Reveal About the Future of Commerce hbr.org/2026/04/rese... by Mark J. Greeven, Fabrice Beaulieu and Wei Wei.

O'Reilly - Generative AI in the Real World: Aishwarya Naresh Reganti on Making AI Work in Production with Ben Lorica and Aishwarya Naresh Reganti - www.oreilly.com/radar/podcas...

New paper: Beyond the tool view of AI: Intelligent technologies and the emergence of new epistemic regimes - Anastasia V Sergeeva, Paul M Leonardi, Samer Faraj, 2026 journals.sagepub.com/doi/full/10....

Just published on FOSS Force: FunOS: Ubuntu LTS with JWM and No Snaps fossforce.com/2026/04/funo...

OpenAI: We’re burning money like the Joker. A miracle needs to happen for us to turn a profit Microsoft: Please please use our AI systems, we’re teetering on the edge here Anthropic: I wonder what’ll kill us first: lawsuits, regulations or model collapse Media and universities: AI is here to stay

New paper in Science, Technology, & Human Values by David Ribes and Eric Monteiro journals.sagepub.com/doi/abs/10.1... Titled: More Instrument than Data

Happy International Women's Day!

Brits would decisively support rejoining the European Union if a new referendum were held, according to the latest polling from YouGov, with clear majorities emerging across all age groups – especially among younger voters www.thelondoneconomic.com/politics/bri...

This is very good.

Very interesting read by Dario Amodei, the CEO of @anthropic.com www.darioamodei.com/essay/the-ad...

Introducing: The Security issue. Security can mean national defense, but it can also mean control over data, safety from intrusion, and so much more. Explore the way technology, mystery, and the universe itself affect how secure we feel in the modern age when you read the full issue. ⬇️⬇️⬇️

Trying to exit my Bluesky-slumber with a research-related post :) Here is a link to the study I am most excited and proud of this year. There are expectations for rationality & human-likeness in the AI field. Is that a problem? See the paper to find out :) onlinelibrary.wiley.com/doi/10.1111/...

Artificial Intelligence at Work: An Integrative Perspective on the Impact of AI on Workplace Inequality, by Arvind Karunakaran, Sarah Lebovitz, Devesh Narayanan and Hatim A. Rahman | Academy of Management Annals journals.aom.org/doi/abs/10.5...

This really can’t be pointed out enough

MIT following Harvard's lead here

Bravo Harvard

Harvard President, Alan M. Garber: "We have informed the administration through our legal counsel that we will not accept their proposed agreement. The University will not negotiate over its independence or its constitutional rights." www.harvard.edu/president/ne...

Half of Cambridge's European students have vanished. The Finnish undergraduates disappeared when fees were introduced. www.varsity.co.uk/news/29471

I just attended a conference in the US. We had one speaker detained and two others turned away at the border.

Applications open for #ECIS2025 Junior Faculty Consortium in Jordan. Deadline: April 16. The co-chairs include Andrew Burton-Jones, Amany ElBanna and me. ecis2025.eu/events/junio... It promises to be a fantastic event so please do apply soon.

Privacy isn’t about hiding—it’s about trust and control over our own information. In her #SXSW 2025 Keynote, Signal President, @meredithmeredith.bsky.social breaks down why our privacy matters.

Huge crowd at Stand up for Science

How can Brexit Britain navigate Trump's World? New post on my Brexit & Beyond Blog. With the rapid deepening of a dangerous global divide, the UK faces hard choices. Could the latest proposals for UK-EU 'mutual recognition' be part of the answer? chrisgreybrexitblog.blogspot.com/2025/02/how-...

The NYT asked me to write an op-ed on Musk and the Treasury and then took a pass. I hope it will still get read.

This white paper calls for collective action among developers, policymakers, and civil society to establish shared standards and implement solutions that balance open sharing with responsible governance. https://opensource.org/datagovernance

Google had created that pledge as a concession to workers campaigning against Pentagon drone AI contracts in 2018. Here's some of that history, that can put the chaotic events into context to make sense of what is new, what is old. jacobin.com/2018/06/goog...

Hugging Face researchers aim to build an ‘open’ version of OpenAI’s deep research tool

📢 𝗡𝗲𝘄 𝗽𝗮𝗽𝗲𝗿 alert! We document how 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗔𝗜 𝗶𝘀 𝗿𝗲𝘀𝗵𝗮𝗽𝗶𝗻𝗴 𝗱𝗲𝗺𝗮𝗻𝗱 𝗳𝗼𝗿 𝗳𝗿𝗲𝗲𝗹𝗮𝗻𝗰𝗲𝗿 𝘀𝗸𝗶𝗹𝗹𝘀. 💼 💻 📊 We analysed 𝟯𝗠+ 𝗷𝗼𝗯 𝗽𝗼𝘀𝘁𝘀, used embeddings to cluster them into 𝟭𝟬𝟬+ 𝗳𝗶𝗻𝗲-𝗴𝗿𝗮𝗶𝗻𝗲𝗱 𝘀𝗸𝗶𝗹𝗹𝘀, and LLMs to classify them as s𝘂𝗯𝘀𝘁𝗶𝘁𝘂𝘁𝗮𝗯𝗹𝗲, 𝗰𝗼𝗺𝗽𝗹𝗲𝗺𝗲𝗻𝘁𝗮𝗿𝘆, or 𝘂𝗻𝗮𝗳𝗳𝗲𝗰𝘁𝗲𝗱.

China’s cheap, open AI model DeepSeek thrills scientists www.nature.com/articles/d41...

I am running an online seminar series @esadebsls.bsky.social for scholars of information systems, innovation and management. The speakers include @asusarla.bsky.social Emma Vaast, Likoebe Maruping, Frank Nagle and more. Please reach out to me if you would like to be added to the mailing list.

This new piece with @toubiana.bsky.social and Trish Ruebottom in @sciam.bsky.social, explored how Trump’s appeal mirrors the strategies of voyeuristic businesses, like reality TV or social media influencers. 🧵 How Reality TV Helps Explain Trump’s Success www.scientificamerican.com/article/how-...

He accidentally did a Nazi salute... TWICE. While talking about the future of Western civilization. He is who we think he is.

Justice is coming... eventually. ⚖️ #MartinLutherKingDay2025

Earlier today I wrote a short and personal piece about the struggles of life in academia as an early-career and untenured researcher. It seems to be resonating with folks over at LinkedIn, so thought I'd share it here as well: www.linkedin.com/pulse/darker... #academicsky

Opinion: Bluesky’s open protocol allows people shape their own experience of social media. That's nothing short of revolutionary—and we need to protect it.

OSI Director Sayeed Choudhury introduces the Open Forum for Artificial Intelligence, a Carnegie Mellon University led partnership that will bend the arc toward human-centered, responsible, and ethical AI https://www.youtube.com/watch?v=dDbjlN0jPFc

Francois Chollet -- "o-models FAR beyond classical DL" @fchollet.bsky.social www.youtube.com/watch?v=w9WE...

“Incredibly enough, my career seems to be going better now that I’m older and wrinkled and fatter. So go figure it out.” Isabella Rossellini talks to Michael Schulman about aging, feminism, and why she never got plastic surgery.

“Facebook launched its fact-checking program after Donald Trump was first elected. Now that resistance appears out of fashion, the company has decided to end it.” The head of our fact-checking department writes about Meta’s decision to stop checking facts.

A few thoughts on Community Notes. - The bridging design of community notes has lots of merit, and the crowd-sourced (vs centralized) model has trust advantages. But Meta has not yet committed to open sourcing their data+code+changes as X has. Or given details of their implementation plans. 1/

A Happy New Year from London! Let's hope there is peace around the world in 2025, especially in the Middle East. www.youtube.com/watch?v=EBLb...

Oxford in December fog

The 10th edition of the Empirical Management Conference has just started! We have a great line up of presenters and papers: www.hbs.edu/faculty/rese... Just started a list of participants of current and past editions, ping me if you want to be added.

It is such a stupid and obvious market failure that nobody has made a consumer AI LLM product that is 1. trained on consensually-acquired material 2. powered with renewable energy 3. genuinely open about its weights and models. Just achieving these things and being creator-friendly would be massive.

Two tenure-track positions in AI at the School of Information Sciences, UIUC. Artificial intelligence is broadly construed here to include AI policy, governance, ethics, AI for cultural heritage, &c. Review of applications will start Jan 6th. Please RT and share! #MLSky #NeurIPS

Good read! New paper in Strategy Science: Theory Is All You Need: AI, Human Cognition, and Causal Reasoning, by Teppo Felin and Matthias Holweg pubsonline.informs.org/doi/full/10....