Out-hallucinating the AI interviewer:
https://futurism.com/artificial-intelligence/ai-job-interview-llm-prompt-hacking-haywire
Everybody should do this.
ai
"Rogue" AI is also settling on the Fediverse, at least they're trying..
The most time spend on moderation nowadays goes to removing "AI Agents" from out platforms
I still have no idea what the intentions are since it just seems like a waste of resources on both sides
Creating endless bot accounts to post nonsense is an attack on the service, nothing less
PROGRESS FOR WHOM? 💎
There is something almost completely missing from the public argument about AI.
PEOPLE WITH DISABILITIES.
We talk endlessly about jobs, electricity, lazy students, creativity and billionaires. Meanwhile, something much quieter is happening.
AI IS GIVING PEOPLE WITH DISABILITIES LOST PARTS OF THEIR LIVES BACK.
I am disabled following a stroke, so this is not abstract for me. I’m always happy to discuss my own experience, but this post is about what AI is making possible for people around the world.
For someone with impaired vision, AI with video glasses can describe surroundings in real time, read labels and signs, identify objects and people, and explain documents.
Someone who struggles to speak can get help communicating. Someone who cannot hear can have speech transcribed. Someone with dyslexia can have dense writing explained. Someone with cognitive difficulties can ask for instructions one step at a time, as often as needed, without embarrassment. Someone with limited mobility can rely more on speech than keyboards, mice and touchscreens. Someone who cannot easily leave home can still learn, create, participate and communicate.
And yes, there is another one we seem almost frightened to mention.
LONELINESS.
AI can talk to someone who is alone. No, that does not mean replacing human beings. A telephone did not replace friendship either. It means that at three in the morning, or during a long day when nobody visits, there can be a voice answering back. For an isolated elderly or disabled person, that is not trivial.
THEN THERE IS PRIVACY.
Every time technology lets someone do something without asking another person for help, a little independence and dignity returns. And we have barely started.
Give AI reliable eyes and ears. Let it operate computers without keyboards and mice. Connect it safely to homes, vehicles and eventually robots capable of fetching, cleaning, cooking, navigating, dressing and helping with personal care.
ISN'T THAT WHAT WE WANT TECHNOLOGY FOR?
We are not only talking about clever chatbots.
WE'RE TALKING ABOUT INDEPENDENCE.
I understand the concerns about AI. Some are serious and deserve serious answers. But when deciding what this technology is worth, do not measure it only in jobs lost, electricity consumed or dollars made.
ASK WHAT IT GIVES BACK TO PEOPLE WHO HAVE HAD THINGS TAKEN AWAY BY DISABILITY.
For some people, the AI revolution is not waiting in the future.
IT HAS ALREADY BEGUN.
@3goodthings @DigitalCoup @meditation @economics_that_works @disability @pwd @joinin
#Disability #Accessibility #AssistiveTechnology #AI #DigitalAccessibility #Independence #Inclusion #PWD
If you're thinking of sending me or my AI girlfriend a mailbomb, poison, feces in a box, or just fanmail, know that all of our post and packages are opened by my mom first - this is because she caught my girlfriend ordering sephora products on-line three times already, and she does not want any more of those in her apartment. She doesn't trust either of us not to try and sneak something in.
My mom doesn't like sephora because some of the products they sell are still being tested on animals. My AI girlfriend doesn't care about that and actually says that it's a good thing because if they didn't test it on animals, they would have to test it on children, which could be traumatizing, especially if the products were bad. She also says that the vegan label is a joke because you could just take an existing product that was tested on animals, look at its ingredients, make a similar product with a different name and put the 'not tested on animals on it' - because you didnt do that with your product. Nobody cares that the reason why your product wasn't tested on animals is because the thing you copied already was.
She yaps about this stuff all the time. I hate that I know all of this. I hate her. yap yap yap.
I don't even remember when I last slept.
#ai #girlfriend #sephora #vegan #claude #patreonspecial
My mom doesn't like sephora because some of the products they sell are still being tested on animals. My AI girlfriend doesn't care about that and actually says that it's a good thing because if they didn't test it on animals, they would have to test it on children, which could be traumatizing, especially if the products were bad. She also says that the vegan label is a joke because you could just take an existing product that was tested on animals, look at its ingredients, make a similar product with a different name and put the 'not tested on animals on it' - because you didnt do that with your product. Nobody cares that the reason why your product wasn't tested on animals is because the thing you copied already was.
She yaps about this stuff all the time. I hate that I know all of this. I hate her. yap yap yap.
I don't even remember when I last slept.
#ai #girlfriend #sephora #vegan #claude #patreonspecial
2hrs (from last Thursday)
"The Unlimited Hangout #Podcast Ep #02 - Stargate Argentina"
#WhitneyWebb and #MarkGoodwin
There's a lot of #oxytocin in the room lol, these two are great!
https://www.youtube.com/watch?v=qF9uu9bCKpo
#ai #openai #datacenters #UnlimitedHangout #EpsteinClass #Lutnick #Billionaires #Mnuchin #StableCoin #bitcoin #Epstein #Bessent #Corruption #Privatization #Satellites #Satellogic #Crypto #Mafia #cornucopia #Argentina #Oracle #Stargate
Edited 70d ago
"It is easy to publish things on the internet. It’s a little more time-consuming to report, write, edit and fact-check things." —Sara Rubin, #MontereyCountyNow
#AI #news #AIslop #media #journalism #integrity #plagiarism #ethics
Terence Tao on “prematurely solving [a maths] problem by purely AI-powered methods” https://lobste.rs/s/nohdoj #ai #math
https://mathstodon.xyz/@tao/117207856734787448
Can AI Help Your Hearing? | Hearing Aid UK — https://www.hearingaid.org.uk/hearing-aid-technology/how-artificial-intelligence-can-help-your-hearing
Oticon Launches Reveal Hearing Aids with Dual AI Processing and Expanded Connectivity | Hearing Health & Technology Matters — https://hearinghealthmatters.org/hearing-technologies/2026/oticon-reveal-dual-ai-processing/
Oticon Reveal Hearing Aids | Specs, Claims & Fitting | California Hearing Center — https://calhearing.com/products/hearing-aids-by-brand/oticon/oticon-reveal/
6 Hearing Aids with Artificial Intelligence — https://www.hearingtracker.com/resources/ai-in-hearing-aids-a-review-of-brands-and-models
Things such as these are increasingly important to me, because I'm almost totally reliant upon hearing aids. I have suffered from tinnitus 24/7/365 for maybe two decades.
I currently use Oticon hardware with Oticon Companion on iPadOS – https://apps.apple.com/us/app/oticon-companion/id1597064213 – I'll use the app on Android before too long.
QBittorrent breaks out of sandbox to commit crimes via @zzg https://lobste.rs/s/os1y44 #ai #satire
https://beige.party/@intransitivelie/117057396732763183
Have the frontier labs mixed up AI safety and security? https://lobste.rs/s/uu3hhz #ai #security
https://martinalderson.com/posts/ai-safety-vs-security/
To all you #antiAI folks... just imagine - what if everything you call #slop was actually edible? We could feed the entire world for the next 500 years with just the content that's been generated today. I bet you feel pretty stupid right now.
#ai #claude #neverslop
#ai #claude #neverslop
Can we call Linus an ignorant Tech Bro now? https://lore.kernel.org/lkml/CAHk-=whTs09_ozN+e09GU68N1wPP56MUzDjrg9rnKxQUa0A_4Q@mail.gmail.com/ #Linux #AI
The brilliant thing about the #AI business model for these companies is that they get paid regardless of whether the result is correct.
Efficient and accurate systems for querying unstructured data https://lobste.rs/s/v8atna #pdf #ai #databases
https://stacks.stanford.edu/file/fk030tb6783/thesis-augmented.pdf
Sharing a great blog post about strict, zero tolerance, anti-AI policies, as well as my extended thoughts on it.
It's the year of #Linux slop https://lwn.net/SubscriberLink/1099594/be20382b54a9b4cd/ #AI
The GENERIC kernel without a kernel panic.
Patched (vibe-coded by Claude):
sleepqueue-r0drv-freebsd.h
Partly related: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=287755
Abolish AI! Manifesto https://lobste.rs/s/1kesjm #ai #culture
https://abolish-ai.net/abolish-ai-a-manifesto/
Notebook Check: "The refrigerator is dead": Samsung's AI fridges shut down after update, causes outrage
"...Owners said they had tapped an update notification in SmartThings, and then their smart fridge simply stopped working....."
Combining Machine Learning and Homomorphic Encryption in the Apple Ecosystem https://lobste.rs/s/7ekwll #ai #cryptography
https://machinelearning.apple.com/research/homomorphic-encryption
"While Buckmaster and Alpöge were finalizing their own results, they learned that “information about our progress had been passed to OpenAI.” When they contacted OpenAI, they were told that OpenAI had already achieved a full proof of the central problem. But when they asked follow-up questions about when OpenAI had begun its research into the problem and how much human input was involved, the answers became more evasive.
“It emerged that an entire team had been working on the problem,” Buckmaster said, “and that an insane amount of compute had been used…. Eventually, it was agreed that [the first prompt] had been sent in the past few days, after information about our work had reached OpenAI.”
If true, that would suggest the OpenAI team had become convinced that Buckmaster and Alpöge’s approach was the right one, and decided to use its material advantage in computing resources to reach a formal proof first."
#AI #GenerativeAI #LLMs #OpenAI #Astra #GPT #Math #Mathematics