Honored to be selected as the 2026 Husky 100 by the University of Washington.
Grateful for my many mentors and friends who have pushed my work in research and beyond, in directions I wouldn't have explored on my own, and steered me to opportunities I'm still growing into.
Onwards!
#Husky100
“Take a moment to think before you dive in. That’s the best advice for Google Photos users, as the company confirms its latest update can scan all your photos to “use actual images of you and your loved ones” in AI image generation.”
www.forbes.com/sites/zakdof...
For a project I started nearly 17 years ago, today I submitted a revised manuscript, responding to a R&R decision from a journal. Pretty good feeling! 🥂 I'll add that the work genuinely improved for all the time it took to get here #slowwork
Thank you. Are you attending?
Got an accepted parallel talk and a poster @ic2s2.bsky.social. See you in Burlington, Vermont, July 28-31, 2026 iA!
(Parallel talk) Detecting and Characterizing AI-Mediated Scientific Code
(Poster) From Job Titles to Jawlines: Using Context Voids for Probing Generative AI Systems
#CSS #SciSci
kind of a poetic justice..
Watching AI "feel" frustrated with its own reasoning is the most relatable thing, and also weirdly satisfying…
Boulder, CO — The Location of ICSSI 2026
The abstract submission window for ICSSI has been extended by a week to April 6, 2026 @ 11:59 PM AoE!
Find guidelines, templates, and submission link here:
icssi.org/guidelines/
See you in Boulder June 29 - July 1, or come on June 28 for the hackathon!
⚗️🔭 How do we spark new scientific discoveries? Why do some breakthroughs seem accidental?
The new book 'The Engine of Scientific Discovery' by CPNSS Research Associate Alexander Krauss is for anyone who wants to understand how we make discoveries.
➡ Link to the book: doi.org/10.1093/oso/...
This.
Much of the "AI Scientist" community focuses on an instrumentalist rather taylorist approach to science. What about the goal of "knowing" and "understanding" the world?
1/15
#ScienceOfAIMediatedScience #ScAISci #AI4Science
A screenshot of Google search with verge headlines rewritten as slop Lego Computer Brick - Engineer James Brown Aug 20, 2022 — James Brown has brilllantly brought classic Lego computer bricks to life by outfitting them with a tiny OLED screen, processor, ... Y/ The Verge https://www.theverge.com› Al> Tech, Reviews: "Cheat on everything" Al tool Apr 23, 2025 - Two 21-year-old Columbia University dropouts are proposing a new $5.3 million twist on the concept: use their Al tool Cluely to "cheat on everything." Read more The Verge https://www.theverge.com» » Microsoft›Al› Tech: Copilot Changes: Marketing Teams at it Again 22 Aug 2024 — Microsoft is getting ready to rebrand its Copilot for its business Al efforts. It's part of an effort to push Microsoft 365 Al tools. The Verge https://www.theverge.com› Wearable Science Tech: Dexcom Stelo hands-on - OTC Continuous Glucose Monitor Aug 26, 2024 - The $99 Dexcom Stelo is an over-the-counter continuous glucose monitor designed for Type 2 diabetics who don't use insulin.
Google is now screwing with the 10 blue links in traditional search and rewriting headlines - including ours - to be the worst kind of slop. This sucks so bad www.theverge.com/tech/896490/...
Also has a coffee or chai dependency, depending on the region.
Finally got to try this zero token architecture. Sure, UI is confusing and very poor documentation, and takes wayy longer to process,..but somehow more accurate? and less hallucinations?..
idk feels suspicious tbh!!
Anthropic doing what it does best: hype this as large-scale qualitative study.
Qual research is not evaluated by scale but by depth, context, and human interpretation. Stop assuming that “hand-wavey-ness” is a qual problem just because qual work doesn’t look like large-N quantitative analysis.
Here's other issues highlighted by AI most of which are valid concerns.
How does Anthropic's Claude characterize it's Interview study?
"It's a large-scale opt-in survey of Claude users with open-ended questions, analyzed quantitatively by Claude."
If only they used AI to review their work.
Some more useful context
x.com/divy93t/stat...
What a joke.
Use AI for survey or interviews or whatever. Whether the method is epistemically sound or not for an analysis is a different debate. But call it what it is.
This is just large scale text elicitation or survey with *shallow* analysis because you did not “deeply engage” with anything as a human. AI just did it for you. Scaling like this hollows qualitative analysis.
Calling it the “largest qualitative study ever” means you don’t know what qualitative research is actually trying to do. Generalization is not a qual problem. It is a quant problem.
Anthropic doing what it does best: hype this as large-scale qualitative study.
Qual research is not evaluated by scale but by depth, context, and human interpretation. Stop assuming that “hand-wavey-ness” is a qual problem just because qual work doesn’t look like large-N quantitative analysis.
Very cool!!
I saw this on X, and it made me think. On one hand, I see the point that many critics may just be critics for the sake of it, impressing one another with how cool they are with how unimpressed they are of tech and AI; but on the other I wonder: isn't *discourse* a desired product of science.
1/8
Idk it's also kind of oxymoron-ish to see the argument for genuine-ness of love for something within the context of technology that most likely -will- make you less genuine in the traditional sense of genuine-ness.
So performance can and does happen both ways; we are all performing here. So what? Hypists hype. Critics critique. Both can be shallow. The important point should be if the discourse is surfacing something new and worth discussing. That's the quality we should care about.
In all, intentions of people are hard to judge. Why people engage in criticism could just be their genuine view or could be performative or often both. And these discussions of intention are somewhat moot. How would you show someone actually *loves* something over them performing that *love*?
In fact, science as an institution can be seen an adversarial collaboration.
So yes criticism will be shallow, performative, or fashionable. That happens in every field. But dismissing criticism misunderstands how knowledge ecosystems and science works. Like, we have a whole system of "peer review" around science part of which is to provide healthy criticism.
