See tagged statuses in the local BookWyrm community
⚡ Researchers Just Found Something Extremely Alarming About AI's Power Usage
｢ As detailed in a new paper, researchers from the open-source AI platform Hugging Face found that the energy demands of text-to-video generators quadruple when the length of a generated video doubles — indicating that the power required for increasingly sophisticated generations doesn’t scale linearly ｣
https://futurism.com/future-society/ai-power-usage-text-to-video-generator
Developers: what are the biggest frustrations when developing generative AI systems?
Hallucinated facts, lack of sources, bias in training data, or all of the above?
The Wikidata Embedding Project is launching on 1 October to help reduce these issues.
If current-generation LLM-based chatbots can drive people to commit crimes or even take their own lives, what do you suppose Neuralink would do to people?
The first thing I said to the person who suggested to me that human brains might be directly hooked to a computer, whenever that was, was "everyone will go insane immediately". I still believe that, but now we're beginning to see evidence.
#AI #GenAI #GenerativeAI #LLM #Neuralink #tech #dev #HCI #BrainImplants
For my latest interview for Stop Gen AI's Anti Gen AI Heroes, I interviewed @tante a hacker, a luddite, and a communist.
We have so much in common! But he definitely codes way better than I can.
Anti Gen AI Heroes: tante moves slow and fixes stuff
https://stopgenai.com/anti-gen-ai-heroes-tante-moves-slow-and-fixes-stuff/
Share, share!
Learning a new AI-related term: workslop.
"The insidious effect of workslop is that it shifts the burden of the work downstream, requiring the receiver to interpret, correct, or redo the work. In other words, it transfers the effort from creator to receiver."
https://hbr.org/2025/09/ai-generated-workslop-is-destroying-productivity
#workslop #AI #slop #GenAI #technology #news #TechNews #study
Had a colleague that had a problem, and "even ChatGPT had to give up".... he showed the error message, 30 seconds searching not looking at (wrong) AI answers, found the github issue and solution.
It's not that I hate AI (well......) - but I hate what it does to people.
This is not a hot take. #science #academics what do you do when fate drops you on a manuscript where a part (in my case the code) has been written using #GenAI and IT SHOWS? I have run, edited, refactored and validated this pile of nonsense but I do not have the resources to write it anew (we started at ca 2000 lines, after my cleaning about 500).
Should our AI statement be “Parts of the code have been generated using ChatGPT without the consent of all authors”? #AcademicPublishing
The context is the field of #paleontology so journals have no or vague requirements
#research #academia
Well-observed article in n+1 magazine
"Large Language Muddle -- It’s OK to be a Luddite!"
https://www.nplusonemag.com/issue-51/the-intellectual-situation/large-language-muddle/
I will give a webinar "Demystifying AI" targeted at non-technical people, primarily working in the third sector or consultants working for the third sector.
It's scheduled for Thursday 18 Sept, at 11:00 UK time, on Zoom.
If you are interested, let me know, I can DM you a link to the registration.
To be clear: I will be explaining why it is bad.
The only hallucinating happening with LLMs is when you, the user, look at the output and think it is real, or true, or correct; or that the LLM is answering a question or following a command. It's not unlike looking at a movie set and believing you're looking at a real town.
#AI #GenAI #GenerativeAI #LLM #Copilot #Claude #Gemini #ChatGPT
🤦♂️ Canadian ‘education report calling for ethical AI use contains over 15 fake sources’ https://arstechnica.com/ai/2025/09/education-report-calling-for-ethical-ai-use-contains-over-15-fake-sources/ #genai #ai #chatgpt #tech #education
*** Important #PSA for #writers and #authors with copyrighted books that may have been stolen by #Anthropic, the #AI company, to train their #LLM.*** There is going to be a final settlement. #SFWA sent out an email about this you will have received if you are a member (I am). The settlement will be substantial.
SFWA Website for Details: https://www.sfwa.org/2025/08/15/press-release-august-15-2025/
Register here: https://www.anthropiccopyrightsettlement.com
Link to legal documents: https://storage.courtlistener.com/recap/gov.uscourts.cand.434709/gov.uscourts.cand.434709.362.0_5.pdf
#writer #author #writingCommunity #writersOfMastodon #llms #genaI
I like to poke LinkedIn once in awhile with an "AI" critique to see what I can stir up. One reason I do this is to keep an eye on the changing form of the booster rhetoric. Nowadays a lot of folks respond to critique with some form of "today's LLMs are bad but tomorrow's will be amazing", the true believer/quasi-religious response with a touch of false humility for flavor. Yesterday I got a "AI critics are just as bad as AI boosters" false dichotomy, which by my read was a variant of the "AI critics are hysterical and irrational" with the twist that the speaker was suggesting that boosters are too. That felt new-ish to me. Granted, the hubristic "we're the smart guys in the room, you should do what we say" framing is ancient in the tech industry. Suggesting the boosters are also not the smart guys in …
I like to poke LinkedIn once in awhile with an "AI" critique to see what I can stir up. One reason I do this is to keep an eye on the changing form of the booster rhetoric. Nowadays a lot of folks respond to critique with some form of "today's LLMs are bad but tomorrow's will be amazing", the true believer/quasi-religious response with a touch of false humility for flavor. Yesterday I got a "AI critics are just as bad as AI boosters" false dichotomy, which by my read was a variant of the "AI critics are hysterical and irrational" with the twist that the speaker was suggesting that boosters are too. That felt new-ish to me. Granted, the hubristic "we're the smart guys in the room, you should do what we say" framing is ancient in the tech industry. Suggesting the boosters are also not the smart guys in the room is an interesting move because it's an attempt to go meta. Neither the boosters nor the critics are the smart guys in the room; the smart guys in the room are actually the ones who can see that (and so you should do what they say, which is more LLMs always).
#LinkedIn #AI #GenAI #GenerativeAI #AgenticAI #LLM
