Me, Anand, at the end of my dissertation defence, with my supervisor Prof. Samer Faraj
Oh hey I'm a doctor doctor*!
Looking forward to what comes next, as an Assistant Professor at the Department of Digitalization at the Copenhagen Business School!
*Only in Germany, and only if I'm in a play dramatizing an Alexander McCall Smith novel.
Found on LinkedIn. What fresh hell is this.
Yet another cycle of market manipulation.
It is art of the deal, we just got the beneficiaries wrong.
A few people have asked for the syllabus from my grad seminar on Generative AI for social science -- just posted it here:
statmodeling.stat.columbia.edu/2026/03/10/n...
Janteloven might be my favourite word/concept I've learned in a while.
Its also utterly fraudlent.
A review by an LLM fined tuned by scraping published texts is in no way comparable to having the original author review the paper, not cognitively, not processually, no aesthetically.
Some people peaked in high school. Now they're looking to impress high schoolers to relive their glory days.
Great resource, thanks!
I wonder if you might consider expanding scope to Academy of Management divisions like @aom-tim.bsky.social , @aomstr.bsky.social , and @aom-omt.bsky.social , as well as publishers like @journals.sagepub.com
So Copenhagen looks like it could be fun.
"Private equity ownership structurally diffuses moral responsibility for organizational decision making," he said, looking at nothing in particular.
Boopgate is the most Canadian thing ever.
Erstwhile peaceful people weirdly aggresive when on ice.
*shakes fist at sky*
Using an LLM-based app to do qualitative research is like using an RPG when you needed a scalpel. Its imprecise, and an unconscionable waste.
For sampling, processing, summarizing, or surfacing themes, there are ALWAYS computationally lighter approaches than using a RAM guzzling behemoth.
A small statement of principle:
AI augmentation should not be seen as a path to excellence. Excellence, rather, should be a pre-condition for AI augmentation.
In knowledge intensive work, this is how we preserve rigor and reflexivity.
Thank for coming to my TED talk.
As you plan for the 2026 conference circuit, know that Calgary is a) gorgeous in May, and b) a short drive from Banff. Hint hint.
Academics on BlueSky, contribute to Blue Sky Thinking at the Administrative Sciences Association of Canada conference.
Qual/Tech folks: the TIM division beckons
Reflecting on recent experience with public sector bureaucrats and my latest field study:
Discretion is anathema to automation, even the fancy “AI” version. In other words, situated judgement is very costly to encode.
Safe: People who can productively bend the rules, and be recognized for it.
Why is it that the most fun part of the PhD (writing the monograph) happens at the same time as the least fun part (applying for a job)? One of life's little ironies.
Based on my preliminary understanding of internet culture, there’s basically no way to say two words and not refer to Benedict Cumberbatch. Pneumatic Cauliflower.
RIP, prince of darkness.
I had similar thoughts when social media was emerging.
Here is a doing that is being done at a massive social scale and there’s an essential quality check component that is missing.
This has implications for both the technical correctness and the aesthetic qualities of what is generated.
Many (most?) genAI users cannot effectively evaluate what they have generated.
It depends on the user and the task. But a good chunk of all genAl tasks are sufficiently specific to require expert or esoteric knowledge to evaluate.
“Did it do a good job?”
The chatGPT (or equivalent) user who can answer that question, can do the job without chatGPT.
A cautionary tale about genAI sycophancy. Treat it like a hotshot analyst. Especially one that produces exactly what you want, in record time. Impressive, no doubt, but requiring considerable oversight and real quality checks.
I'm curious to see how the annual meeting this year in Copenhagen will unfold.
Will Americans show up? Will international scholars in America (who risk not being able to re-enter) do so as well? What about scholars in European universites?
For some, Canada (e.g. ASAC) might offer an alternative.
Respectfully disagree. I like negative spaces in storytelling (see Andor S2 for a well-executed example). Not everything needs to be explained "in canon".
I’m not ashamed to admit that I log out, then look.
Next year, ASAC is going to be held in Calgary, May 22nd-25th 2026.
Alberta is gorgeous this time of the year, and Banff is like an hour and a half drive away from Calgary. Hint hint.
