SpaceX Alleges a Chinese-Deployed Satellite Risked Colliding with Starlink science.slashdot.org/story/25/…
SpaceX Alleges a Chinese-Deployed Satellite Risked Colliding with Starlink - Slashdot
"A SpaceX executive says a satellite deployed from a Chinese rocket risked colliding with a Starlink satellite," reports PC Magazine: On Friday, company VP for Starlink engineering, Michael Nicolls, tweeted about the incident and blamed a lack of co…science.slashdot.org
Francisca Sinn
in reply to Gerry McGovern • • •“He found that at least 11 of 21 citations in the first chapter could not be matched to known academic papers. The analysis also suggested that 8 of the 11 citations in chapter 4 were untraceable.“
#AiSlop #ClownShow
Gerry McGovern reshared this.
Gerry McGovern
in reply to Francisca Sinn • • •Francisca Sinn
in reply to Gerry McGovern • • •Yes. It’s ridiculous. Yes, and yes.
It’s also shocking to me to see how many people know of these nonsense outputs, but then brush it off as outlier issue (a random funny thing that happened that one time) and have their default belief that the output is valid unshaken. It's wild.
reshared this
Gerry McGovern reshared this.
Gerry McGovern
in reply to Francisca Sinn • • •Matt Mascarenhas
in reply to Gerry McGovern • • •Gerry McGovern reshared this.
GeofCox
in reply to Matt Mascarenhas • • •@miblo @fsinn
I came across this on mastodon too. A post linking a blog piece I thought misrepresented a political philosopher, so I checked the citations - which made no sense. I replied to the post - and the author said they were compiled using AI.
The minimum we need is a warning that AI has been used for such things - so we know not to waste time taking the results seriously.
Matt Mascarenhas
in reply to GeofCox • • •David Nash
in reply to Francisca Sinn • • •I know, right? I managed to inoculate myself against the hype almost from the start by asking ChatGPT 3.5 questions about something I knew well, and getting unambiguous confabulated bullshit in response. I then saw plenty of other confabulated bullshit in other topics, all of which I could easily verify was bullshit. My own take on all that — all in the first few months of ChatGPT’s public release — was “it gets numerous easily verified facts wrong, so it is *not trustworthy* on anything”. It was astonishing to see so few other people I know come to the same conclusion.
Gerry McGovern
in reply to David Nash • • •@dpnash
Like you, I used it once for some serious research and was not impressed.
@fsinn
Francisca Sinn
in reply to Gerry McGovern • • •@dpnash I’ve never used it, didn’t bother to check for any serious research in law because I already knew it was weighing blog posts from lawyers equally with blog posts from non-lawyers , nevermind “reddit-lawyers”, and it can’t be trusted to deliver complete and accurate results. I don’t use it for drafting because my clients expect 25 years of experience, not common answers. Resulting value to me: zero.
Thought you’d find this interesting -
mas.to/@fsinn/1157323023249102…
Francisca Sinn (@fsinn@mas.to)
Francisca Sinn (mas.to)Gerry McGovern reshared this.
Martin Rundkvist
in reply to David Nash • • •@dpnash @fsinn
One of the first things I tried was asking the bot to write a brief introduction to myself. It got a lot right -- after all, there's a long Wikipedia article. Then it made up a fictional book with a strangely unspecific title and claimed that I'd written it. (No, not the Necronomicon.)
#llm
Maxi 12x 💉
in reply to David Nash • • •Rene Gat
in reply to Francisca Sinn • • •Rene Gat
in reply to Rene Gat • • •PeachMcD
in reply to Gerry McGovern • • •Gerry McGovern reshared this.
tlariv
in reply to PeachMcD • • •@theonion's writing is better. It's like they have copyeditors.
@gerrymcgovern @theonion
Androcat
in reply to Gerry McGovern • • •Sensitive content
This is wholly unsurprising.
People writing ethics guides on a method are generally in favor of the method.
Otherwise the guide would just be one word: "Don't".
And everyone who is in favor of (LLMs posing as) "AI" is a lazy nincompoop.
Gerry McGovern
in reply to Androcat • • •�
in reply to Gerry McGovern • • •Gerry McGovern reshared this.
Squads
in reply to Gerry McGovern • • •gabriele renzi
in reply to Gerry McGovern • • •Gerry McGovern reshared this.
the roamer
in reply to Gerry McGovern • • •"You can't make this stuff up."
Well, you and I can't. The LLM can. 😀
#StopTheAICorruption
Helen LH
in reply to Gerry McGovern • • •Gerry McGovern reshared this.
Richard Wonka
in reply to Gerry McGovern • • •QED.
Brilliant.
Graeme 🏴
in reply to Gerry McGovern • • •Gerry McGovern reshared this.
Zoidberg For President
in reply to Gerry McGovern • • •LR
in reply to Gerry McGovern • • •staringatclouds
in reply to Gerry McGovern • • •Apparently you can make this stuff up
As evidenced by the AI in the article
Gerry McGovern reshared this.
Neil Moffatt
in reply to Gerry McGovern • • •Gerry McGovern reshared this.
James 🦉 #FBPE 🇪🇺
in reply to Gerry McGovern • • •Gerry McGovern reshared this.
gbsills
in reply to Gerry McGovern • • •MHLoppy
in reply to gbsills • • •,K PkNikku4211xX
in reply to MHLoppy • • •I could not even fathon having to pay for LLM slop articles.
eruwero
in reply to Gerry McGovern • • •skua
in reply to Gerry McGovern • • •Gerry McGovern reshared this.
Ray Gulick, he/him/wtf 🇺🇦 ❌👑
in reply to Gerry McGovern • • •@AIscam
Gerry McGovern reshared this.
arclight
in reply to Gerry McGovern • • •Three plus or minus five
in reply to Gerry McGovern • • •Why is "fake" in quotation marks?
If the evidence shows it, why waffle?
Gerry McGovern reshared this.
Matthew Abbott
in reply to Three plus or minus five • • •Three plus or minus five
in reply to Matthew Abbott • • •@matt
Experts differ on color of sky. "Blue" say some.
They are disowning responsibility for any judgement, even as they are making those judgments implicitly by editorial decisions. It is cowardice. The sky is blue. The citations are fake. Any disagreements are either disingenuous or picking nits.
Matthew Abbott
in reply to Three plus or minus five • • •Three plus or minus five
in reply to Matthew Abbott • • •@matt
So why the quotation marks? They convey that the statement is subject to opinion. Now, if they're reporting eye-witness accounts, ok, the evidence comes from a source that can't be checked. But in this case the citations are clearly visible and checkable. They could check them and find them fake. Why not?
Journalism has to accept responsibility for stating the truth, not just citing people who state things.
No Exceptions
in reply to Gerry McGovern • • •Ben Royce 🇺🇦 🇸🇩
in reply to No Exceptions • • •@Bartok
i was going to say
"fake citations found in AI ethics guide" sounds par for the course
Alberto Cottica
in reply to Gerry McGovern • • •Gerry McGovern
in reply to Alberto Cottica • • •James
in reply to Gerry McGovern • • •The Sleight Doctor 🃏🍉
in reply to Gerry McGovern • • •I wonder if it contains any passages on the "ethics" of claiming to have authored a textbook you didn't write and haven't even fact-checked.
Because that would be ironic.
Gerry McGovern reshared this.
Alexander 😷
in reply to Gerry McGovern • • •Debora Weber-Wulff
in reply to Gerry McGovern • • •Gerry McGovern
in reply to Debora Weber-Wulff • • •Julian ♪🌻🥥🌴♫
Unknown parent • • •LLMs don't "learn" the way humans do. Look at how AI kept denying the death of Pope Francis because it was trained on data up to a certain point and could not integrate new information at all.
@dpnash @fsinn @gerrymcgovern
,K PkNikku4211xX
in reply to Gerry McGovern • • •Felicitas Macgilchrist
in reply to Gerry McGovern • • •Gerry McGovern reshared this.
Jordan Biserkov
in reply to Gerry McGovern • • •Gerry McGovern reshared this.
Anselm Bühling
in reply to Gerry McGovern • • •This is the same publisher who last year brought out a machine translation of a book on Marx‘ Capital from English into German where the Chapter heading „Population and Rent in Capital“ was rendered as „Population and (Housing) Rent in the Capital City“.
norden.social/@ichichich/11332…
ichichich (@ichichich@norden.social)
norden.socialGerry McGovern reshared this.