Boosted by ChrisWere@toot.wales ("Chris Were ⁂🐧🌱☕"):
infobeautiful@vis.social ("Information Is Beautiful") wrote:
Six ways to divide England *linguistically*
by @starkeycomics
Boosted by ChrisWere@toot.wales ("Chris Were ⁂🐧🌱☕"):
infobeautiful@vis.social ("Information Is Beautiful") wrote:
Six ways to divide England *linguistically*
by @starkeycomics
Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
davidgerard@circumstances.run ("David Gerard") wrote:
AI bot network Facebook buys AI bot network Moltbook
the dumbest AI bubble acqui-hire this week
https://www.youtube.com/watch?v=%5FUGaMXRawDU&list=UU9rJrMVgcXTfa8xuMnbhAEA - video
https://pivottoai.libsyn.com/20260311-ai-bot-network-facebook-buys-ai-bot-network-moltbook - podcasttime: 5 min 46 sec
https://pivot-to-ai.com/2026/03/11/ai-bot-network-facebook-buys-ai-bot-network-moltbook/ - blog post
Boosted by jwz:
SnoopJ@hachyderm.io wrote:
hearing gullible 20-somethings say "this technology DOES have good use-cases, like in medicine for example…"
is going to turn me into the fucking Joker
Boosted by jwz:
mawhrin@circumstances.run ("flere-imsaho 🇺🇦") wrote:
it is not a good day to be a conscientious user of modern technology, no.
i just read the financial times article (archive) on the use of confabulation machines in war planning, specifically the scope of use in target planning, and it's bad, bad, terribly bad.
even if it is what we suspected anyways, knowing the lack of ethics of the tech companies and the bloodthirstiness of hegseth & al.
there is no hope for this industry.
Boosted by jwz:
coral@empty.cafe ("Coral (bleached era)") wrote:
the FT reports today that the US Military is using an AI kill chain, built by Anthropic and Palantir.
They use it to launder inadequate evidence into kill decisions no-one can be blamed for, to perpetuate a terror program where it's impossible to tell who counts as a "combatant". And they will kill you, your children, your neighbours, and your children's entire school, for reasons you will never know.
Boosted by jwz:
r_flash@mastodon.r-flash.eu ("Adam P") wrote:
Boosted by jwz:
jpm@aus.social ("I love this, so I") wrote:
@zzt for real, I wonder what they’d do if they encountered U+130BA ?
db@social.lol ("David Bushell ☕") wrote:
i am far too busy to be wasting hours trying to score the god run on this! please block my IP
i swear i saw 39/40 once but i keep fat fingering when my phone decided to sleep
https://www.keithcirkel.co.uk/whats-my-jnd/?r=ANkgKP%5F%5F89fU
db@social.lol ("David Bushell ☕") wrote:
noted: idle thoughts on the element
https://dbushell.com/notes/2026-03-12T06:37Z/— can't see myself ever using this thing
Boosted by zkat@toot.cat:
mshelton ("Martin") wrote:
Signal recently released another grand jury subpoena which, as always, basically had nothing interesting to turn over in response. For the requested phone numbers they could only provide the account creation timestamp. https://signal.org/bigbrother/district-of-columbia/
Boosted by slightlyoff@toot.cafe ("Alex Russell"):
hollie@social.coop ("Hollie") wrote:
Seen today on my bike ride up to the market.
Boosted by slightlyoff@toot.cafe ("Alex Russell"):
Z@jorts.horse ("zeevacation@gmail.com") wrote:
logging in like
slightlyoff@toot.cafe ("Alex Russell") wrote:
Turned out I had a (unpaid) Grammarly account.
Emphasis on "had".
jscalzi@threads.net ("John Scalzi") wrote:
Look, I strongly believe we will get through this and we will be better than whatever the fuck this is right now, and I will work to get us there and so will lots of you, but I resent that we will still have to live through all the time between here and there and I wish we were all there already.
kevinevans@hachyderm.io ("Kevin") wrote:
Did a code review today where somebody who obv used AI created a button styled like a link to open a page with window.open... instead of, ya know, creating a regular link. 🫠
It bugs me that this person didn't seem to review the generated code AT ALL and instead pushed that shit to me to review. What's worse is half the team rubber stamps PRs, so I know there's some nasty code making it to prod.
dysfun@treehouse.systems ("gaytabase") wrote:
left pack
gagging to be made into an npm library
RE: https://sfba.social/@MLNow/116213005270150379
So Danny Bluejeans is now driving around town, hopping out of the car, and having his non-uniformed security detail assault random people.
And then they get their asses kicked.
Normal mayor. Normal city.
dysfun@treehouse.systems ("gaytabase") wrote:
Artificial German Insertion (AGI)
fromjason ("fromjason.xyz ❤️ 💻 ✍️ 🥐 🇵🇷") wrote:
Marcia Roy
I suspect some of you may say this is where I'm stretching this theory. But Marcia is where I think the theory is the strongest. Or, at least, she's what made me want to explore the theory more.
Marcia represents Israel. She has Logan's ear in a way that none of the kids have it, and they're suspicious of her influence.
Even more interesting, I think the show hints at the possibility that Marcia might've been an Israeli spy...
Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
wordshaper@weatherishappening.network ("Dan Sugalski") wrote:
@soatok and you can even pronounce that “no” like “the ROI on collecting that data is negative” because it is for basically everyone. The suits often need you to speak with an accent that way so they can hear you clearly.
Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
petrillic@hachyderm.io ("Chris Petrilli") wrote:
RE: https://furry.engineer/@soatok/116212999782522947
People: You need a union for representation and collective bargaining.
SWE: No! We are special. We are powerful. Hear us roar! RAWR!
People: Great, tell your boss no.
SWE: Oh no, we could never do that, we'd lose our jobs.
Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
canacar@ioc.exchange ("Can Acar") wrote:
@soatok 100% agree that "No" is a great answer for both security and privacy related problems. It is good to always question what problem they are trying to solve. If nothing else, it is one less thing to worry about if you manage to shut it down. It is not always an acceptable answer, however, if they are already (committed to) doing it. Then you would have to go into damage reduction mode.
Sometimes, "if I don't get involved they will ship it anyway" realization leads to proposing such alternatives.
dysfun@treehouse.systems ("gaytabase") wrote:
i've got a knackered old combine harvester and you can hotwire it for free
Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
cliffle@hachyderm.io wrote:
@soatok respecting privacy is the ultimate privacy-enhancing technology
soatok@furry.engineer ("Soatok Dreamseeker") wrote:
Last thing: When I said "No" is a Privacy-Enhancing Technology, I didn't just mean an opt-out.
I mean the engineers growing a fucking spine and telling their boss, "No, we shouldn't collect this data in the first place."
Boosted by jsonstein@masto.deoan.org ("Jeff Sonstein"):
Rasta@mstdn.ca wrote:
"There is no flag large enough to cover the shame of killing innocent people" - Howard Zinn.
soatok@furry.engineer ("Soatok Dreamseeker") wrote:
During the coffee breaks and dinner conversations, everyone I talked to about these things echoed my frustrations.
In 2024, a speaker from Intuit spoke about their distributed key generation protocol. It involved multiplying a number by a hash. They did not elaborate on whether that's just a bigint operation or an elliptic curve group operation. @sophieschmieg was like, "Why would they do that? What if they set it to zero?" and the backchatter was full of "Why are the tax people rolling their own crypto?"
So, like, I'm not super worried about adtech rotting the RWC community.
Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
r0k wrote:
@soatok most of the problems in this area stem from a fundamental misunderstanding of consent
soatok@furry.engineer ("Soatok Dreamseeker") wrote:
At an earlier track, one of the invited speakers suggested using Fully Homomorphic Encryption to allow folks to have private conversations with an AI chatbot for therapy.
My mind was instantly filled with news stories of OpenAI and self-harm. Lawsuits from grieving families.
Are they deeply out of touch?
Or was it just "hmm, what do people want privacy for? I'll just throw a bunch of hypothetical examples of things FHE would be good for without interrogating them deeply"?
Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
S-Config@core.s-config.com wrote:
@soatok "No" is also Ecologically friendly. as it take maybe 1 packet to say "No." and not an endless CPU cryptographic reach-around to tickle the users balls because you refuse to accept "No." and instead put a "Maybe later." increasing energy usage for everyone.
See, this is why I can't be a speaker at these events.