cfiesler

cfiesler, 1 month ago to random

In honor of discovering that I've apparently had a Wikipedia page for a few months, I decided to tell the story of why I'm not a Wikipedian (a story I told Jimmy Wales lol) and this seems like the kind of thing that might interest you all: https://www.instagram.com/p/C5i1r68Lyk-/

Though TL;DR if you don't want to watch a four minute video, my attempt to increase coverage of children's literature on Wikipedia was met with exactly the kinds of gatekeeping acrimonious jerk behavior you might expect.

cfiesler, 9 months ago to random

One piece of grad school application advice I've been giving is to consider sending a draft of your statement of purpose to your letter writers and basically saying: I thought that having this information about my motivations and background might help in drafting your letter, though also if you have any feedback for me I would of course welcome it.

This seems like a clear win-win to me! Do you think this is good advice?

cfiesler, 2 months ago to random

We have a peer review crisis in computer science, and recent evidence of ChatGPT written peer reviews is just more evidence. Meanwhile, here are things I've been hearing re: benefits of gen AI:

(1) We can draft papers faster!
(2) We can do research faster for things like qualitative analysis of large data!
(3) We might not even need to run human subjects studies, since we can use AI to simulate participants!

Anyway, expect the "way too many papers being submitted" problem to get way worse.

cfiesler, 1 year ago to random

So Mashable published a story a few days ago about Bluesky's TOS that is absolutely mind-blowing levels of wrong. Like, it cites this TOS provision and then say: "In plain English that means: we own everything you post." No it absolutely does not mean that. It says non-exclusive license right there! WTF? Like Bluesky's TOS might very well have issues, but this is not one of them. And then goes on to claim that DMCA safe harbor doesn't apply to Bluesky. ???? BAFFLED.

These terms might have downsides for Bluesky itself But owning user content, and being able to transfer rights to that content, may mean BlueSky is not protected by the "safe harbor(opens in a new tab)" provisions of the Digital Millennium Copyright Act (DMCA). Under the DMCA, safe harbor protection goes away if a company has the "right and ability to control(opens in a new tab)" copyright-infringing content and can potentially profit off of it. That means these terms leave Bluesky vulnerable to costly copyright infringement litigation, which would sap the fledgling app's potential.

cfiesler, 5 months ago to random

As a reminder, if you are a teacher, do not use AI detectors. This piece mirrors what I've heard from a LOT of students on social media (panicked students commenting on my AI videos, with no reason to lie). In my opinion, whatever utility you're getting out of a way to catch cheating is not worth the risk of even one false positive. Especially the potential for systematically biased false positives. https://www.thedailybeast.com/ai-written-homework-is-rising-so-are-false-accusations

cfiesler, 1 year ago to random

We ( @bkeegan @MattNicholson @blakeley ) are researchers at University of Colorado Boulder conducting a survey about social media platform migration. If you are at least 18 and have used Twitter and/or Mastodon you are eligible to share your experiences with leaving (or choosing not to leave) Twitter and/or joining (or choosing not to join) Mastodon. You can optionally provide your email address to enter a drawing for a $75 gift card. Please boost! Survey link: https://cuboulder.qualtrics.com/jfe/form/SV_3BLXjcmBmKpyR2S

cfiesler, 1 year ago to random

peak AI hype is James Cameron thinking he needs to wait to see where real AI is going before making a new Terminator movie https://movieweb.com/james-cameron-working-on-new-terminator-movie/

cfiesler, 8 months ago to random

The comments on my women in computing videos just keep on coming, and today it was someone who says he's been in IT for 30 years and has "never, ever" met a woman who has shown any interest in computers.

I seriously didn't know how to respond except to say that it's okay, not everyone meets women. Because what other explanation could there possibly be?!

cfiesler, 1 month ago to random

I seriously don't understand how it's possible that I just see news story after news story that could have been made up by a group of undergrads I asked to imagine things that might obviously go wrong with generative AI. https://themarkup.org/news/2024/03/29/nycs-ai-chatbot-tells-businesses-to-break-the-law

cfiesler, 1 month ago to random

In anticipation of the imminent end of my sabbatical (and not much travel this month), April is going to be a content creation push for me. Any suggestions for topics? This will mostly be short form video (Instagram/TikTok) though I might be persuaded to a blog post or two. But I think my question for you is: What do you think people need to know? About AI or other tech ethics, IP law, internet stuff, HCI, women in computing, academia... etc.? I know some of you might have good ideas. :)

cfiesler, 9 months ago to random

Academics, I beg you. Please put copies of your published papers on your website, arxiv, whatever. Not only because it's great for the broader world beyond academia (who can't get behind paywalls at all) but it just took me 15 minutes to try to figure out how to log in to some particular database to get a copy of a paper. I don't even like the 3 minutes it takes to log into the ACM DL versus copying a title into google. Follow copyright rules but you can almost always post SOMETHING somewhere.

cfiesler, 4 months ago to random

Have we decided yet what the appropriate gender-neutral substitute for sir/ma’am is? Because especially when I’m back in Georgia (maybe I regress to being ten years old) I sir/ma’am service workers constantly out of habit.

cfiesler, 10 months ago to random

and then a 700 billion dollar company launched a social media app without the ability to add alt text to images

cfiesler, 2 months ago to random

It is #SIGCSE2024 - the big ACM computer science education conference! There years ago I made this YouTube video about my thoughts about ethics integration across computing curriculum - I have a lot more thoughts on this since then, but if you've never seen me give a talk on this topic, this is a good introduction! https://youtu.be/LYRKqnLeIDo

cfiesler, 1 year ago to random

I'm actually kind of surprised that Open AI wrote this down. Like, straight up stating that though it's so so important that we regulate hypothetical technology that might harm people in the future, we should definitely not regulate the technology that exists today and is harming people in ways we can actually see... is quite a bold strategy. https://openai.com/blog/governance-of-superintelligence

cfiesler, 1 month ago to random

Here's today's academic hot take. The way we run conferences is absolutely wild. Can you imagine if a company had 100% employee turnover every year and all they had to rely on was (if you're lucky) some documentation, or having to reach out to the person who previously had your job which you then feel bad about because they don't work there anymore?

Also the fact that anyone thinks professors should be event planners.

cfiesler, 7 days ago to random

re: Scarlett Johansson I think it's interesting that Mira Murati told NPR that they didn't pattern the ChatGPT voice off of anyone and that she doesn't even know what Johansson's voice sounds like. Because that could totally be true.

Anyone else think Sam Altman might have just asked for input into voice casting, and (after Johansson turned them down) chose the actress whose voice sounded most like hers? Which, you know... illegal? No. Creepy? Yes.

cfiesler, 6 months ago to random

Saw a social media post from a grad student saying that a paper reviewer said in their review that they could tell some of the paper was written by AI. OMG DON'T DO THAT. I really hope this isn't the new version of reviews that suggest a paper was written by a non-native English speaker. Uuuugh.

If you must critique the writing, then actually critique it. (If you think it was written by AI, what about that writing is bad? e.g. is it stiff? Inaccurate?) Never assume cause...

cfiesler, 2 months ago to random

Hey folks, does anyone know of any work so far about (or people working on) neurodivergence and AI detectors? There's been some great writing with anecdotal evidence, but it seems like really low hanging fruit that I'd love to see more on. (I'm not asking because I'm thinking of doing it - I just want to see someone else's work. :) )

cfiesler, 1 year ago to random

I have a new social media landing page! https://casey.prof/

cfiesler, 9 months ago to random

I posted a video on Instagram about women in computing, and the misogynist trolls are out in force! But I've discovered that the best response to the (dozens of) comments from men I've received along the lines of "CS isn't toxic for women" or "no one is keeping women out of CS" is: "Oh it's so great to see a woman in CS who has had such a positive experience comment on my post!"

cfiesler, 1 month ago to random

The first rule of etiquette on Reddit is "remember the human" - this should apply to researchers, too! This blog post is a summary of our (myself, @michaelzimmer @profprof @sarahgilbert and Naiyan Jones) recently published paper (in PACM-HCI as part of the ACM GROUP conference) about research ethics for Reddit and public internet content more generally. https://cfiesler.medium.com/how-to-remember-the-human-recommendations-for-ethical-reddit-research-8fe34e30e21d

cfiesler, 8 months ago to random

I knew this was going to hit mainstream news. I seriously still feel like I'm being punked. I must know: Men of Mastodon, how often do you think about the Roman Empire??? https://www.washingtonpost.com/lifestyle/2023/09/14/roman-empire-trend-men-tiktok/

cfiesler, 1 month ago to random

On a post that has absolutely NOTHING to do with this, two people have been arguing in my comments for three days about whether spreadsheets are databases. The thread is now almost 50 comments.

cfiesler, 1 month ago to random

I would really love if every time I post something about AI bias on the internet I didn't have to (a) defend my credentials ("do you even code bro?"); (b) justify that yes, I know how AI works, and yes, I know what training data is; (c) clarify that more than one thing can be a problem at the same time (e.g., tech AND society); and/or (d) deal with people who insist racism has to be intentional.