Assoc Prof | Information Science
Ex-investigative reporter and editor
Crisis informatics nerd
Makes good trouble
Tell your dog I said "hi"
wendy norris
Loading...
It will never stop being funny to me that the street fighter subreddit has a much better, well reasoned ai policy vis a vis literally all universities.
Let’s take a look at how Apple’s auto captioning is working on Elmo
CJ Fogler
Faisal فيصل
Somebody get these little guys a one-way ticket to DC.
Make Algae Great Again!
"[These] promises are all false. In this article we explain how and why. [Our] alternative approach disavows attempts to automate science—using machine learning or otherwise—and instead centers scientists’ own cognition." -- me and @olivia.science
THIS all damn day.
If these brainiacs really thought they had some legit use for LLM/LGM-powered generative text app, then make your case. Hiding behind ethics-washing and high-minded "responsible" tech use ain't it.
Thinking about LLM use in research: Assembling text or images for models should be held to the same standards as other data collection methods. Otherwise, IRB would clobber us and rightfully so for using non-consensual data. ChatGPT, Gemini, Copilot, Claude, etc. don't meet that ethical standard.
Friends in DC: We need your help.
Post the patches/insignias of local LEs that you see carrying out unlawful detention and arrests for crimes like touching reflecting pool water. US Marshals deputized local LEs to serve in DC. Our county sheriffs and city PD chiefs need to get an earful from us.
Let's go!
See also
www.techpolicy.press/ais-catastro...
Trump is now encouraging a Sunni fundamentalist Syrian militia to invade Lebanon to attack an Iranian backed Shia fundamentalist militia. What could go wrong?
wendy norris
wendy norris
Iris van Rooij 💭
wendy norris
wendy norris
wendy norris
Sasha Costanza-Chock
For many, AI introduces the suspicion that self-investment is inherently a losing proposition, writes recent graduate Evan Liu.
there is no such thing as "responsible use of generative ai" in research
Handful of detainments here by the reflecting pool. They are being done a variety of officers, including Monroe County Sheriff, Oklahoma City, Oklahoma State Police, US Marshals, a county in Texas I can’t make out, and a county in New York I can’t make out.
The most dangerous thing about AI is not that the machines might one day think. It is that we are being trained, by men with enormous financial gains riding on our acquiescence, to stop thinking for ourselves.
