Last spring I taught a graduate course built on the idea that 𝘧𝘭𝘶𝘦𝘯𝘤𝘺 𝘸𝘪𝘵𝘩 𝘈𝘐 𝘪𝘴 𝘢 𝘳𝘦𝘴𝘦𝘢𝘳𝘤𝘩 𝘴𝘬𝘪𝘭𝘭, not a novelty: 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗔𝗜 𝗳𝗼𝗿 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵—all the course materials are now public. Take it, use it, adapt it for your own students or your own research. barbagroup.github.io/mae6291-genai/

The Journal of Open Source Software, JOSS, is embarking on the search for its next editor-in-chief. Founder EIC Arfon Smith is stepping down after a whole decade of incredible service. Interested in filling those big shoes? Apply by June 19, 2026: blog.joss.theoj.org/2026/05/call...

JOSS is seeking a new Editor-in-Chief | Journal of Open Source Software Blog

After a decade of building and running JOSS, our founding Editor-in-Chief is stepping down. We're seeking a new EiC to lead editorial direction for the Journ...

blog.joss.theoj.org

Arfon Smith@arfon.bsky.social · 3mo ago

After 10 years, I'm stepping down as Editor-in-Chief of @joss-openjournals.bsky.social. A few thoughts on the decade: www.arfon.org/stepping-dow... The EiC call: blog.joss.theoj.org/2026/05/call... #opensource #openscience

I just published open materials from our "AI Academy" for engineering faculty. Session 4 tackled the big shift: AI now does much more than answer your questions. It can go away, do research, and come back with a deliverable—faculty need the framework to understand this.

At #JupyterCon, I saw the future of AI collaboration for data science and beyond—and it's not in a chat window. Abigayle Mercer and Zach Sailer demoed an experimental new version of Jupyter AI and it changes the plot for multi-agent systems. A thread with my thoughts… 🧵

𝗖𝗼𝗺𝗲𝘁 𝗿𝘂𝗻𝘀 𝗰𝗶𝗿𝗰𝗹𝗲𝘀 𝗮𝗿𝗼𝘂𝗻𝗱 𝗝𝘂𝗽𝘆𝘁𝗲𝗿 Friends, I watched an AI agent complete my students' coding exercises in real-time, and I'm shook 🤯 I tested Perplexity's Comet browser on a Jupyter notebook with eigenvalue problems I assigned to my class… (video at the bottom of this thread)

I just rejected a paper because it was pointless in the age of AI. For the first time, I recommended a manuscript be rejected because its content was so easily replicable by an AI model with deep research capabilities. #AcademicPublishing #AIinResearch

🧵 Fall 2025 faculty: I wrote 7 pages on what you need to know about AI before classes start in a week or two—it's posted as PDF in the ‪@figshare.com‬ service under CC-BY: link below… Here are the highlights: 1/8

Good benchmarks allow a user to choose the best method for a particular application. “But the first question is, what do we mean by ‘better’?” Fascinating story by @drmichaelbrooks.bsky.social on benchmarking in #AI. Thx @labarba.bsky.social for the suggestion! 🧪 www.nature.com/articles/d41...

Is your AI benchmark lying to you?

Artificial intelligence models are too often assessed against flawed goals — a stumbling block for progress.

nature.com

Had a hallway conversation at #SciPy2025 about #GenAI in coding education → one week later, a (partial) solution is shipped in @googlecolab.bsky.social 🤯 The magic of conference serendipity never gets old. Longer post with story on LinkedIn: www.linkedin.com/feed/update/...

A week has passed since I got back from the SciPy Conference in Tacoma, WA, delighted as usual by the experience. | Lorena A. Barba

A week has passed since I got back from the SciPy Conference in Tacoma, WA, delighted as usual by the experience. I had skipped it last year, weary of crossing the whole continent for the event. But I...

linkedin.com

Presentation slides for my talk yesterday at #SciPy2025 Barba, Lorena A. (2025). Embracing GenAI in Engineering Education: Lessons from the Trenches. figshare. Presentation. doi.org/10.6084/m9.f...

I did a hands-on tutorial for easily creating a static website using #Jekyll and #GitHub Pages. No local installations—everything is done directly on @github.com! Follow along to: – Fork a Jekyll theme – Deploy with GitHub Actions – Edit the content – Solve issues with #GitHubCopilot (link in reply)

I just released a preprint reporting on my experiment with generative AI in my Engineering Computations course. TL;DR: It was a mess. The biggest issue? The "illusion of competence." #GenAI #education #edtech

Starting now! So excited for our inaugural #GWOSCON Dean Geneva Henry welcoming everyone with optimism about the role of #OpenSource in our university. Inspiring!

CURIOSS@curiossorg.bsky.social · last yr.

🍀 Best of luck to #CURIOSS member, @david-lippert.bsky.social, and all involved in the inaugural #GWOSCON next week! 🏆 What a great tagline from @labarba.bsky.social : "Free to attend, free t-shirt, free lunch, free coffee ☕️ And of course, free and open source software 🚀" ℹ️ bit.ly/4imRlDZ

There's still time to book your free ticket for GW OSCON from March 24-25 next week. Register: go.gwu.edu/oscon #opensource #academicOSS #openscience #openresearch

March 2, 1867. CHAP. CLVIII. — An Act to establish a Department of Education. "…for the purpose of collecting such statistics and facts as shall show the condition and progress of education in the several States and Territories."

I'm at the "Uncertainty Quantification and Trustworthy AI Algorithms for Complex Systems and Social Good" workshop" at the Institute for Mathematical and Statistical Innovation, IMSI, in Chicago. Should be fun! (Funded by NSF.) #UQ #AI #TrustworthyAI www.imsi.institute/activities/u...

📣New paper alert: "Verification and Validation for Trustworthy Scientific Machine Learning" We propose 16 recommendations to establish trust in #SciML models through rigorous V&V practices adapted from computational science: arxiv.org/abs/2502.15496 #MachineLearning #ComputationalScience

Verification and Validation for Trustworthy Scientific Machine Learning

Scientific machine learning (SciML) models are transforming many scientific disciplines. However, the development of good modeling practices to increase the trustworthiness of SciML has lagged behind ...

arxiv.org

I'm finally getting around to editing videos from online teaching during pandemic times (to remove dead times and student names and likeness), and uploading them to YouTube. This is for my "Engineering Computations" undergraduate course: #Python functions #EngineersCode #Jupyter

The George Washington University, where I work, has joined a lawsuit as a co-plaintiff in challenge of the federal action to slash payment of facilities and administration (a.k.a. "indirect") costs for NIH research grants. www.aau.edu/newsroom/pre...

Statement of Association of American Universities, Association of Public & Land-grant Universities, and American Council on Education Regarding Their Legal Challenge to the Administration’s Cut to Lif...

AAU, ACE, and APLU have jointly filed suit in U.S. District Court in Massachusetts, along with a number of impacted research university co-plaintiffs, seeking to halt the NIH's proposed cut to life-sa...

aau.edu

The deadline to submit a regular or lightning talk for the inaugural GW OScon has been extended until Monday. Are you in the Washington DC area and love #OpenSource software? Apply to give a talk and join us in March! We have great keynote speakers lined up, too. Check out: go.gwu.edu/oscon

CURIOSS@curiossorg.bsky.social · 2y ago

#GeorgeWashingtonUniversity #OSPO are hosting their inaugural #opensource conference, GW OSCON from March 24-25 and it promises to be fantastic! There's still time to apply to speak. ℹ️https://ospo.gwu.edu/open-source-conference-gw-oscon #academicOSS, #openscience, #openresearch

DeepSeek is the perfect illustration for why you don’t want vendor lock-in. The university CTOs that signed a contract for ChatGPT Edu at $1.4million must feel sorry right now.

Our systems of peer review & publishing are crumbling. Predatory publishers, exorbitant costs, unchecked bias, and AI are eroding trust. It’s now more important than ever for scientists to be able to affordably and easily share their findings. 2/

Word of the Day is ‘uhtcearu’ [ucht-kay-aru, with the 'ch' as in the Scottish ‘loch']: Old English for ‘the sorrow before dawn', when you lie awake in the darkness and worries crowd your mind.

We absolutely need to deliberately teach students how to use a variety of AI tools, and so-called "AI skills" are real and more difficult than they appear. The temptation to overuse AI to skirt the hard work of authentic learning is strong! Unsupported, students' use of AI is harming them right now.

Simon Willison@simonwillison.net · 2y ago

I don't think students should take classes on how ChatGPT Code Interpreter differs from ChatGPT Canvas - that would be like teaching MS Word specific VB macros in "intro to computers" I do think there's a lot of space for helping teach students how to navigate these utterly weird new AI systems

I'm having way too much fun with AI-generated podcasts of my class materials 😂 — I hope students will enjoy them too! youtu.be/dI8JBHB0kv0

Engineering Computations 1 – Lesson 1 – Interacting with Python

YouTube video by Lorena Barba

youtu.be

"OpenAI is aggressively going after the education market, which it sees as a key area of growth." techcrunch.com/2024/12/05/o... University CTOs beware: we need to avoid vendor lock-in! OpenAI GPT-4o may be the most-used model today, but things are moving fast in this sector.

Lorena A Barba@labarba.bsky.social · 2y ago

A very bad deal for universities: $1.4million price tag annually, for an estimated 10k users, and vendor lock-in. Reminds me of the first years of edX, when universities paid millions to become members of the "consortium" plus annual fees. A high price, resulting in no in-house capacity building.

I tried alphaXiv, from arXiv Labs: the platform for open discussion of #arXiv papers. But instead of commenting on one the latest research preprints, I added an author's comment on one of the first papers coming out of my group, 15 years ago! www.alphaxiv.org/abs/0809.181...

Characterization of the errors of the FMM in particle simulations | alphaXiv

View 1 comments: This was the first paper coming out of my research group investigating the fast multipole algorithm (FMM), with my first ever PhD student. (Felipe Cruz is, fifteen years later, a seni...

alphaxiv.org

Like many professors, I sometimes get bizarre requests from members of the public with an, ahem, scientific interest. This one ranks high on the bizarro scale!

Here's a thought—what if we just, you know, stopped using number of publications as a measure of researcher quality? What if we decided that "publish or perish" was a bad idea from the start? Just at thought. Anyway, fascinating thread here!

Altmetric@altmetric.com · 2y ago

In the years ahead, the tools supporting scholarly misuse of AI will scale as technology does. Increasingly convincing, refined and above all, large in volume, they are a hefty wave in the battle for reality. A wave that has already broken and receded while another readies. 30/31

As an educator, I'm waking up at night in angst. It is _our challenge_ to find ways to convey responsible use of #AI to our students, lest they go out in the world thinking they can one-shot-prompt their way through any task and end up causing some debacle and getting fired. #generativeAI