@katyilonka profile picture
katy ilonka gero @katyilonka
Joined May 2014
  • Tweets 2,345
  • Following 476
  • Followers 985
159 Photos and videos
Pinned Tweet
katy ilonka gero@katyilonka
19 Nov 2024
After a long break from twitter, I plan to take an even longer one. Instead, I'll be on the cozy web [1], and maybe mastodon. Send me an email! Also: I am recruiting PhD students to join me at the University of Sydney. Check my website for details. [1] contraptions.venkateshrao.co…

The Extended Internet Universe

The Internet is now a very complicated place

contraptions.venkateshrao.com
2
20
2,513
katy ilonka gero retweeted
Keyon Vafa@keyonV
15 Aug 2024
Excited to announce the NeurIPS 2024 workshop on Behavioral Machine Learning! Are you working on AI methods that incorporate models of human behavior? Submit a paper by Sept 10! We invite submissions across CS fields and the behavioral sciences. More: behavioralml.org

1
19
89
20,855
I find this extremely hilarious.
Find me on bsky @colin-fraser.net@colin_fraser
23 May 2024
left it running all night
4
1,000
OpenAI wants to build a 'Media Manager' for creators to "tell us what they own and specify how they want their works to be included or excluded." In my research with writers, writers don't trust big tech to manage such a thing--- openai.com/index/approach-to…

Our approach to data and AI

Just over a year after launching ChatGPT, AI is changing how we live, work and learn. It’s also raised important conversations about data in the age of AI. More on our approach, a new Media Manager...

openai.com
1
3
326
They also don't trust academics, publishers, or even agents. They only trust other writers to advocate for their interests. OpenAI says they're collaborating with creators, content owners, & regulators, but it can't be that collaborative when there are such divergent interests.
1
1
254
Creators should be governing themselves, and finding the leverage such that tech companies have to respect their own solutions, not the other way around.
1
221
katy ilonka gero retweeted
Caroline Haskins@car0linehaskins
22 May 2024
New/exclusive: African workers who moderate platforms like Instagram train AI for places like OpenAI (often for >$2/hr) pub'ed an open letter to Biden. They say US tech companies need to stop paying "modern day slavery" level wages & union busting: wired.com/story/low-paid-hum…

The Low-Paid Humans Behind AI’s Smarts Ask Biden to Free Them From ‘Modern Day Slavery’

African workers who label AI data or screen social posts for US tech giants are calling on President Biden to raise their plight with Kenya's president, William Ruto, who visits the US this week.

wired.com
5
216
421
61,952
Just read "The dark side of optimism: Musical dreams, belief, and gambling", and the authors note that similar themes emerge in other high-risk careers like artists more broadly, professional sports and even **academia**. journals.sagepub.com/doi/epu…

1
362
more replies
even if it does feel like far too many research papers are coming out, and people are publishing way more than is reasonable. That, too, may be driven by a sense of scarcity--if I don't publish enough, I won't be able to have a successful career.
1
1
180
If we could get around a scarcity model, and encourage and highlight ways for more researchers to be successful, people might feel less need to rush through their research, and instead do slow work with stronger results.
174
amidst AI news, I have been working through taper 12, a literary magazine for small computational pieces. each piece is just 2kb big, and issue 12 is about tools --- tools to change the world, tools to understand it, as well as literal tools you can use. taper.badquar.to/12/

1
2
6
981
more replies
for some reason I'm drawn to Jackie Liu's difficult and depressing to-do tool: taper.badquar.to/12/to-do.ht…

To-Do

A poem by Jackie Liu for Taper #12

taper.badquar.to
1
133
keep the internet big and wild and weird. hang out in some small spaces. 🌳🌳🌳
125
Load more