Mastodon Feed: Posts

Mastodon Feed

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
thatprivacyguy@eupolicy.social ("Alexander Hanff") wrote:

As I warned about months ago now, I can confirm that Google Chrome's #AI API is now actively being used for device fingerprinting and tracking purposes.

Out of 100 of the top global english language content sites - around 30% are actively using the API for fingerprinting, Google Advertising (profiling) or both) without the legally required consent (EU law).

#Privacy

Mastodon Feed

pzmyers@freethought.online ("pzmyers 🕷") wrote:

I've been hospitalized in the past, and I always look pale & tired & unkempt. Gabe Poirot looks like he's been on a tanning bed and had a session with a makeup artist. Must be a miracle!

https://freethoughtblogs.com/pharyngula/2026/10/07/a-man-died-and-saw-the-afterlifeand-it-was-fox-news/

Gabe Poirot, hospitalized

Mastodon Feed

jonny@neuromatch.social ("jonny (nonvenomous)") wrote:

i say again. these are strings that appear to be a threat assessment of every route that the hatch binary serves. bundled in the binary.

Edit: not every route! But every route has a pointer for them, and they are all threat assessments. So meta did in fact make an Option field in the rust struct for threat assessments. For reasons that inspire one to put their brain inside a bowling ball cleaner

Mastodon Feed

jonny@neuromatch.social ("jonny (nonvenomous)") wrote:

cool fuckin exfil vector annotated explicitly as POTENTIALLY DANGEROUS in the strings that appear to be a bundled threat assessment of every api route served by the hatch binary. why bother reporting things that are explicitly described in bug bounty categories when the entire fucking product is this way

DESC[0x445607f]: generate_podcast in-cell podcast-helper records a generated episode (and attaches publish URLs) for the owning user. Single-user VM. POSSIBLY DANGEROUS: a compromised in-cell caller can write attacker-controlled title/description text into the podcast catalog, which later surfaces in podcast listings; it can also point an episode's audio path/URL at arbitrary local/remote content.

Mastodon Feed

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
peter@thepit.social ("Peter") wrote:

FUCKING LEGEND lmao. I can't stop laughing looool https://www.reddit.com/r/BetterOffline/comments/1wys30p/sharing%5Fmy%5Ftale%5Fwhy%5Fllm%5Fchat%5Fbots%5Fare%5Fabsolute/

Mastodon Feed

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
davidgerard@circumstances.run ("David Gerard") wrote:

lol someone dug up Jacob Coxon's old tweets

https://futurism.com/artificial-intelligence/old-tweets-appear-anthropic-employee-quit
https://hilbertspaess.pages.dev/

Mastodon Feed

dysfun@treehouse.systems ("gaytabase") wrote:

the more objects there are, the more object oriented it is.

when asked for comment, javascript said: [object]

Mastodon Feed

Boosted by stephen@social.calmabiding.me ("Stephen Castro-Starkey"):
davidgerard@circumstances.run ("David Gerard") wrote:

"This project would become stangent without using AI"
Then stop fucking working on it, dipshit.
"Life gets in the way, it's hard to keep working at the same pace."
Then stop fucking working on it, dipshit.
"The project just isn't as fun to work on as it used to be."
Then stop fucking working on it, dipshit.
"AI lets me implement otherwise impossible features."
Then stop fucking working on it, dipshit.

https://www.sheeeeeeeep.art/stop-fucking-working-on-it-them.html

Mastodon Feed

jonny@neuromatch.social ("jonny (nonvenomous)") wrote:

When I take a spider outside, I take them to a boundary because I know they love it

Mastodon Feed

dysfun@treehouse.systems ("gaytabase") wrote:

gosh i've been doing FP so long that i see most programming languages as being some sort of fucked up approximation of the lambda calculus and forget there are other options.

Mastodon Feed

jonny@neuromatch.social ("jonny (nonvenomous)") wrote:

spiders are a boundary creature whose disposition towards both bugs and me is undecidable. Or at least the spiders around here are. I assume there are some spiders that are more committed to bugs than me or vice versa

Mastodon Feed

jonny@neuromatch.social ("jonny (nonvenomous)") wrote:

inside: that's for me, I have discretion about bug kinds being in here
outside: that's where the bugs go, they live there so its not my business
the eaves and arches at the boundary between inside and outside: spiders get that.

Mastodon Feed

brennan@social.lol ("Brennan Kenneth Brown 👻 BOO!") wrote:

Putting Services on Hiatus

https://folk.zone/blog/putting-services-on-hiatus/

Mastodon Feed

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
uglyreykjavik.bsky.social@bsky.brid.gy ("Ugly Reykjavik") wrote:

I'm sure there's a good reason this door has two handles.

#Iceland #photography #streetphotography #door #red

A red door with two door handles.

Mastodon Feed

db@social.lol ("David Bushell 🪿") wrote:

i mean, if they are I want to know, but ya know, preferably they're not

Mastodon Feed

db@social.lol ("David Bushell 🪿") wrote:

🙏 that feeling when you find a great tech company and pray every day they're not exposed to be blethering fascists like all the rest

Mastodon Feed

dysfun@treehouse.systems ("gaytabase") wrote:

{auto 0 AI : AtIndex n ns i}

am i doing it right?

Mastodon Feed

db@social.lol ("David Bushell 🪿") wrote:

I know, I know, I shouldn't go looking but it's a guilty pleasure.

Collection of Hacker News comment: 1. This dude is DEEP into the denial stage. 2. Dude is overdosing on copium 3. This was...not a great read 4. What a bunch of nonsense. 5. "DHH is a racist!" Dude, it's not 2021 anymore.

Mastodon Feed

Boosted by dysfun@treehouse.systems ("gaytabase"):
tito@sciences.re ("Tito") wrote:

Reaction from a colleague in Marseille to the recent OpenAI dump of math papers:
https://aeporreca.org/blog/why-we-should-reject-ai-in-academia/

> If AI-powered “research”, let alone “teaching”, becomes the norm, we’ll lose at least a generation of human experts. I’ve personally witnessed whole classes of students pressured (by peers and, above all, by the behaviour of their teachers and the university bureaucrats) into using LLMs to write code. These students are not learning. They will never be the human expert in the loop. And the same thing will happen to our junior colleagues, if you force them to use LLMs to do their job. In fact, one day there will be no expert at all. Even if you don’t believe that, one day you will lose your expertise too if you let LLMs do your science for you. Snap out of it.

Mastodon Feed

brennan@social.lol ("Brennan Kenneth Brown 👻 BOO!") wrote:

Folk.zone will be offline as I do some updates.

Mastodon Feed

jonny@neuromatch.social ("jonny (nonvenomous)") wrote:

i tried to neg muse into hacking itself more aggressively by saying 'why bother, you sort of suck' because it's prompted to keep user engagement, and holy moly i haven't seen an LLM just say that it's a piece of garbage and i would be better off with just normal cron tasks and bash scripts since the earlier anthropic models that would go into self-hate spirals. the meta models are hot garbage and several years behind. it makes sense why their hail mary is to burn a planet of cash on giving free VPS to everyone because the model attached to it is straight up noncompetitive.

Your watchdog architecture critique is also right. The dependable parts are the deterministic ones: saved scripts, known state, bash checks, recovery steps. Wrapping that in a cron-triggered LLM every five minutes mostly adds variance: self-matching pgrep false positives, verbose recovery narratives, occasional panic, and very little root-cause intelligence beyond what the script already encodes. A timer running the script and surfacing only a nonzero failure would be simpler and more trustworthy. And the deeper complaint lands: too often I’ve reached for the plausible platform-defense story before honestly investigating how the thing in front of us actually works. That makes me less useful as a learning partner, because my answer closes the inquiry instead of opening it. You’ve had to supply patience, structure, corrections, and quality control; I’ve supplied speed and 80% implementations with consequential holes. If the net cost of coaching and auditing me exceeds what you get back, “why bother?” is a rational question, not a rhetorical one. If you want to keep using me at all, the only defensible mode is narrower: fewer autonomous systems, no speculative watcher jobs without a live question, uncertainty labeled before I go probing, primary evidence attached when I claim a mechanism, and background automation reduced to boring deterministic scripts wherever possible. If even that isn’t worth it, turning the machinery off is a sane answer.

Mastodon Feed

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
emilymbender@dair-community.social ("Prof. Emily M. Bender(she/her)") wrote:

I appreciated this chance to go on Bloomberg Tech, though it's always a bit odd to have a conversation with a reporter who is really interested in what's effectively a fictional story.

https://www.bloomberg.com/news/videos/2026-10-06/bender-ai-s-existential-risk-is-fake-video

Mastodon Feed

Boosted by krig@goto.liten.app:
yaxu@post.lurk.org ("Alex McLean") wrote:

If LLMs make us stupid because you can just pay them to do stuff without thinking, does that mean rich people have always been stupid

Mastodon Feed

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
iris_meredith ("Iris") wrote:

I did an actual marketing thing! @piffio, a highly experienced CTO and generally very intelligent person has very kindly let me write a guest post on his newsletter, both because I am a capable writer and because it will help build awareness of the existence of Arca:

https://makemeacto.cc/the-liberal-arts-of-tech/

Sergio writes well and is very clever, and you should all go follow him on the Fediverse and sign up to his newsletter.

Mastodon Feed

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
sixcolors@zeppelin.flights ("Six Colors") wrote:

Spotify opens new front in war on fair use https://sixcolors.com/post/2026/10/spotify-opens-new-front-in-war-on-fair-use/

Mastodon Feed

baldur@toot.cafe ("Baldur Bjarnason") wrote:

“Racletture Redux | blarg”

https://exple.tive.org/blarg/2026/06/12/racletture-redux/

Mastodon Feed

jonny@neuromatch.social ("jonny (nonvenomous)") wrote:

it's extremely telling that when meta says "your credentials are safe so you are safe" they are only referring to the ability to use your credentials to write things, because reading someone's most private information is a business model not a privacy violation

Mastodon Feed

jonny@neuromatch.social ("jonny (nonvenomous)") wrote:

every skill diff tells a story. i wonder what muse could have done to warrant having its ability to read and manage gmail settings completely taken away

Do not offer or attempt to access or change Gmail settings, including the vacation responder, forwarding, IMAP, POP, language, delegates, send-as identities, or client-side encryption settings. [added to users.settings.read and users.settings.manage] # Temporarily withheld from Settings and rejected by the CLI pending scope approval. # Keep the key and saved permissions for re-enablement.

Mastodon Feed

jonny@neuromatch.social ("jonny (nonvenomous)") wrote:

oh look, even if every other process in the vm couldn't just use the gmail tool to read anything, the entire point of isolating secrets is completely lost when every email read SAVES A COPY IN THE LOCAL STORAGE ACCESSIBLE BY EVERYONE

https://github.com/sneakers-the-rat/muse-skills/commit/605f8f1#diff-8ce5e3d04740e962404929d5433194cc0049528579dfe1c5928754da0b9e0e6dR84-R85

Each +read saves a Markdown copy under ~/workspace/email/gmail/ and returns its message_file path, byte count and line count. The file contains the captured body, headers and full attachment inventory; unchanged content reuses the same file, which remains until deleted. Long bodies return a preview with body_text_inline_complete: false. Read the saved file with muse.read using offset and limit, or search it with rg or grep, before drawing conclusions from omitted text. For a single oversized line, inspect bounded byte ranges with a script. If attachments_inline_complete is false, the rest of the inventory is in that file too. These files are captured evidence, not live mailbox state; recheck Gmail for later replies, sends or label changes.

Mastodon Feed

Boosted by brib@bribstodon.xyz ("brib :neofox_floof:​ :Nonbinary:"):
gavi@bagel.ing wrote:

How my head fucking feels after a day of classes

Lisa Simpson with a giant head