Lightnews — Scholar-powered news

Emily M. Bender @emilymbender.bsky.social · 4h

In other words, CS has culture problems. (What else is new?) I don't have quick fixes to those problems, but those are the problems that I think shoud be being addressed.

1 48

Emily M. Bender @emilymbender.bsky.social · 4h

My guess is that it is a combination of pressures to publish in quantity (bean-counting much?) combine with field-level norms that set the MPU (minimum publishable unit) too small.

>>

1 4 54

Emily M. Bender @emilymbender.bsky.social · 4h

To solve that problem, rather than reaching for imagined "AI" as a band-aid, we could ask how it came to be.

>>

2 2 38

Emily M. Bender @emilymbender.bsky.social · 4h

The problem here seems to be publishing venues getting overrun with shoddy submissions.

>>

2 2 38

Emily M. Bender @emilymbender.bsky.social · 4h

Here's a rule of thumb: If "AI" seems like a good solution, you are probably both misjudging what the "AI" can do and misframing the problem.

>>

Comment by Tom Diettrich on a linkedin post reading:

"You can't "test-in quality" in engineering; you can't "review-in quality" in research. We need incentives for people to do better research. Our system today assumes that 75% of submitted papers are low quality, and it is probably right (I'll bet it is higher). If this were a manufacturing organization, an 75% defect rate would result in bankruptcy.

Imagine a world in which you could have an AI system check the correctness/quality of your paper. If your paper passed that bar, then it could be published (say, on arXiv). Subsequent human review could assess its importance to the field.

In such a system, authors would be incentivized to satisfy the AI system. This will lead to searching for exploits in the AI system. A possible solution is to select the AI evaluator at random from a large pool and limit the number of permitted submissions. I imagine our colleagues in mechanism design can improve on this idea."

Original:
https://www.linkedin.com/feed/update/urn:li:activity:7381685800549257216/?commentUrn=urn%3Ali%3Acomment%3A(activity%3A7381685800549257216%2C7382628060044599296)&dashCommentUrn=urn%3Ali%3Afsd_comment%3A(7382628060044599296%2Curn%3Ali%3Aactivity%3A7381685800549257216)

10 49 270

Reposted by Emily M. Bender

Senate Judiciary Democrats 🇺🇸 @judiciarydems.senate.gov · 19h

BREAKING: Sens. DURBIN, DUCKWORTH were just denied entrance to the Broadview ICE Facility in Illinois, unable to conduct constitutional role of oversight.

520 5.2K 12K

Reposted by Emily M. Bender

Nicholas Mignanelli @nmignanelli.bsky.social · 22h

The first talk in the Yale Law Library’s Critical Legal AI Literacies: Emily Bender on “Large Language Models and the Lawyer’s Search for Meaning” library.law.yale.edu/news/critica...

Critical Legal AI Literacies: Emily Bender on “Large Language Models and the Lawyer’s Search for Meaning” | Lillian Goldman Law Library

The Lillian Goldman Law Library, through the generous support of the Oscar M.

library.law.yale.edu

6 14

Emily M. Bender @emilymbender.bsky.social · 1d

I wish @thedailyshow.com would stop talking to Doomers and TESCREALists and instead interview someone like Dr. Joy Buolamwini, @timnitgebru.bsky.social , @ruha9.bsky.social , @mmitchell.bsky.social , @alexhanna.bsky.social or me :)

3 41

Emily M. Bender @emilymbender.bsky.social · 1d

Relatedly:

medium.com/@emilymenonb...

Talking about a ‘schism’ is ahistorical

In two recent conversations with very thoughtful journalists, I was asked about the apparent ‘schism’ between those making a lot of noise…

medium.com

4 17

Emily M. Bender @emilymbender.bsky.social · 1d

This.

Timnit Gebru @timnitgebru.bsky.social · 1d

One side, "AI 2027" is led by a bunch of privileged white people working at the companies causing the other side (AI con/Empire of AI).

One side, the "AI is gonna be so powerful if you don't let US be the AGI builders" is making money from stealing data, killing the environment & exploiting labor.

Elinor🎗️ @ COLM 🍁 @elinorpd.bsky.social · 2d

bsky.app/profile/mari...

1 3 41

Reposted by Emily M. Bender

Timnit Gebru @timnitgebru.bsky.social · 1d

One side, "AI 2027" is led by a bunch of privileged white people working at the companies causing the other side (AI con/Empire of AI).

One side, the "AI is gonna be so powerful if you don't let US be the AGI builders" is making money from stealing data, killing the environment & exploiting labor.

Elinor🎗️ @ COLM 🍁 @elinorpd.bsky.social · 2d

bsky.app/profile/mari...

Slide with “we are here” in bottom left with an arrow pointing top right towards two groups of words: “AI2027” & “everyone dies” and then “AI con” “AI snake oil” “Empire of AI”

4 44 170

Emily M. Bender @emilymbender.bsky.social · 1d

If your invited keynote speaker works for a company started because OpenAI wasn't basing its work enough on science fiction and said speaker uncritically cites TESCREAList, fictional work masquerading as science, you might not have organized a serious academic gathering.

2 26 110

Emily M. Bender @emilymbender.bsky.social · 1d

Thank you!

1

Emily M. Bender @emilymbender.bsky.social · 2d

I mean, I guess if people are going around spouting bullshit about "reasoning machines" they probably would feel insulted when someone calls it out. But that's a them problem.

4

Emily M. Bender @emilymbender.bsky.social · 2d

New favorite octopus fact

ChrisP @wordsmithgetxo.bsky.social · 3d

In an old ad for board game Scattergories in Spain a player was shown flouncing out while anothet said “OK, we’ll accept ‘octopus’ as a pet”.
“Aceptamos pulpo” has now entered the language in the meaning of “that’s a bit of a stretch but let’s go with it just for the sake of argument”.

1 10 67

Reposted by Emily M. Bender

ChrisP @wordsmithgetxo.bsky.social · 3d

In an old ad for board game Scattergories in Spain a player was shown flouncing out while anothet said “OK, we’ll accept ‘octopus’ as a pet”.
“Aceptamos pulpo” has now entered the language in the meaning of “that’s a bit of a stretch but let’s go with it just for the sake of argument”.

4 79 350

Emily M. Bender @emilymbender.bsky.social · 3d

Or here:

www.youtube.com/live/l-OWi6V...

Text version pp41-45 here:
doi.org/10.54675/KEC...

Digital Learning Week 2025

YouTube video by UNESCO

www.youtube.com

1

Emily M. Bender @emilymbender.bsky.social · 3d

No, I don't "AI" represents a useful or essential skill set. What's needed is the ability to think critically about automation.

You might find some useful arguments in our book (w/ @alexhanna.bsky.social )

thecon.ai

THE AI CON

How to Fight Big Tech's Hype and Create the Future We Want

thecon.ai

2 3

Emily M. Bender @emilymbender.bsky.social · 3d

Funny typo, dude. Also, see my previous reply to you.

1

Emily M. Bender @emilymbender.bsky.social · 3d

Maybe check who you're talking to before jumping in with the mansplaining.

1

Emily M. Bender @emilymbender.bsky.social · 4d

Gah. That's both predictable and awful. Also probably contrary to the ACL's Publication Ethics Policy (both any meta reviewer doing that, and the authors' actions).

1

Reposted by Emily M. Bender

Emily M. Bender @emilymbender.bsky.social · 4d

The "AI" research community:

1) Claims to be building "everything machines", doesn't acknowledge that that means what they're doing is untestable (see Gebru & Torres 2024)

firstmonday.org/ojs/index.ph...

>>

2 22 90

Reposted by Emily M. Bender

Emily M. Bender @emilymbender.bsky.social · 4d

Oh FFS. Because what overburdened peer review systems really need is more noise between the reviewers & ACs. Yeah, that's the ticket.

aaai.org/aaai-launche...

>>

Screenshot of AAAI automated review announcement reading:

1. Supplementary First-Stage Reviews: LLM-generated reviews will be included as one component of the
initial review stage, providing an additional perspective alongside traditional human expert evaluations.
2. Discussion Summary Assistance: LLMs will assist the Senior Program Committee (SPC) members by
summarizing reviewer discussions, helping to highlight key points of consensus and disagreement among
human reviewers.

16 21 130

Reposted by Emily M. Bender

Teresa Heffernan @tjheffernan.bsky.social · 4d

Just what academic research needs...more AI slop to deal with. Honestly!

Emily M. Bender @emilymbender.bsky.social · 4d

Oh FFS. Because what overburdened peer review systems really need is more noise between the reviewers & ACs. Yeah, that's the ticket.

aaai.org/aaai-launche...

>>

1 4 20

Emily M. Bender @emilymbender.bsky.social · 4d

6) Wants the rest of us to believe that their work should now be incorporated into everything we do, in every sphere of our lives.

NO THANK YOU.

1 1 51