Mastodon Feed: Posts

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

practicing talking to australians before interview

g'day sheila, didja throw a shrimp on the barbie while you were losing the ashes?

Boosted by jwz:
sfbaykeeper@sfba.social ("San Francisco Baykeeper") wrote:

BREAKING: After a multi-year court battle, we've won a major case over bacteria pollution in the South Bay!

A court has ruled that Sunnyvale and Mountain View are responsible for over 1,200 Clean Water Act violations--and must stop pollution from reaching local creeks: https://baykeeper.org/press%5Frelease/court-rules-that-cities-pollution-of-local-creeks-violated-clean-water-act/

#sunnyvale #mountainview #waterpollution #bacteria #bacteriapollution #cleanwateract #southbay #stevenscreek #calabazascreek

Boosted by jwz:
pikesley@mastodon.me.uk ("CYBER_RISK_MITIGATION_REMINDER") wrote:

INTERCAL: We have keywords like PLEASE and IGNORE because we think it's funny

Anthropic: We have PLEASE and IGNORE *baked into our source-code* because we have no fucking idea how this thing works

Boosted by EmilyEnough@hachyderm.io ("Emily 🏳️‍🌈🏳️‍⚧️"):
caroline@chaosfem.tw ("Caroline Kinlund") wrote:

@EmilyEnough warmly friendly, assertive, broken at least half of the time? checks out

Boosted by EmilyEnough@hachyderm.io ("Emily 🏳️‍🌈🏳️‍⚧️"):
2legged@mastodon.ie ("Claire McNab") wrote:

EmilyEnough@hachyderm.io ("Emily 🏳️‍🌈🏳️‍⚧️") wrote:

Macrumors keeps using this color scheme in its iOS 27 stories, did I miss a report that Siri is a lesbian in iOS 27? Good for her.

jscalzi@threads.net ("John Scalzi") wrote:

Accurate

https://theonion.com/study-97-of-all-sounds-infuriating/

isagalaev ("Ivan Sagalaev :flag_wbw:") wrote:

Eventually the AI industry believes "we will develop similar practices around unread code"[1].

They don't know how, they just believe.

It might even work out, given the amount of talent thrown at it. But right now nobody knows.

But what I'm saying is that "lol this code's a mess" is not such a slam dunk you might hope.

[1]: https://www.heavybit.com/library/article/write-only-code

isagalaev ("Ivan Sagalaev :flag_wbw:") wrote:

Seeing a lot of hot takes on the ongoing #anthropic source leak hilarity bemoaning just how terrible the code is.

Folks, you're missing the point. Their idea is that code is now a low-level implementation detail, it's not for reading. You only program with prompts, and nobody cares how hard it is for agents to read their own code. That humans still need to review it is considered a temporary inconvenience, which they are solving (for now) in the usual way: hiring more bodies. (cont.)

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

yes i will definitely appear human at 1:30am to an observer when it's not even 10pm yet and i want to go to bed

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

wow i can't believe matrix is going to implement reading messages

Boosted by jwz:
dnalounge@sfba.social ("DNA Lounge") wrote:

♬️ OLLIE WRIDE at DNA Lounge tonight: Wed Apr 1, 8pm!
https://www.dnalounge.com/calendar/2026/04-01.html?utm%5Fsource=sp%5Fma
#dnalounge #olliewride #deloreanoverdrive #synthwave #synthpop #livemusic #concert #sanfrancisco

Attachments:

• gifv: 4797317bcc3b3986.mp4

jwz wrote:

A major plot point of *Superman* was that Clark got a garbled answering machine message and didn't realize his parents were Space Nazis who wanted him to take over and eugenics-fuck his way to a new super-race (the plot of *Invincible* and also the plot of Elon Musk). But at the end Supergirl shows up, who left Krypton as a teenager. So couldn't the little blonde übermensch have just... *told* him about her time in the Krypton Youth and at Ku Klux Krypton rallies?
https://jwz.org/b/yk5i

Boosted by adam@social.lol ("Adam"):
neatnik@social.lol ("Neatnik") wrote:

RE: https://social.lol/@neatnik/116328001566273692

There are just a few slots left! :prami_distressed:

denschub@schub.social ("Dennis Schubert") wrote:

time to seize the means of computing?

Boosted by jwz:
malwarejake@infosec.exchange ("Jake Williams") wrote:

ICE was checking IDs at the Atlanta airport today. The.guy said "your DL photo is super light, so facial recognition might not work."

I snapped back "that's so you know not to illegally deport me."

His buddy said "we can find a reason" and NOTHING could be more on-brand for an encounter with ICE.

glyph ("Glyph") wrote:

@aredridel I am taking "measuring your guardrails" to be an empirical ongoing thing, the sort of dashboards and metrics where the sort of infohazard of repetitive plausible-but-bad code would show up, the sort of thing that would be structurally resistant to informal analysis

Boosted by aredridel@kolektiva.social ("Mx. Aria Stewart"):
jackwilliambell@rustedneuron.com ("Jack William Bell") wrote:

@pictureman @Strandjunker

One side of my family has been in the Americas since the 1600s. That side also has native American blood two generations back. IOW? I can trace back my heritage to before the USA existed, in multiple ways.

I say that to say this: I am in favor of easy immigration and citizenship of people's from anywhere. I reject the unstated racism of the anti-immigration crowd – most of whom would not be citizens if their ancestors had come here under the rules they want.

Boosted by jwz:
johl@mastodon.xyz ("Jens Ohlig") wrote:

„By Wednesday morning, Anthropic representatives had used a copyright takedown request to force the removal of more than 8,000 copies and adaptations of the raw Claude Code instructions—known as source code—that developers had shared on programming platform GitHub.“

Because if there’s one thing GenAI companies absolutely don’t take lightly, it’s copyright.

https://www.wsj.com/tech/ai/anthropic-races-to-contain-leak-of-code-behind-claude-ai-agent-4bc5acc7

aredridel@kolektiva.social ("Mx. Aria Stewart") wrote:

@glyph You just said you don't measure your guardrails!

glyph ("Glyph") wrote:

@aredridel and… I do? I may be unfairly assuming you know anything about my previous body of work but I assure you I do a lot of that sort of thing

glyph ("Glyph") wrote:

@aredridel like, "unreleased product info" is _one_ of the things here, but the prompt is quite explicit about being deceptive about being an AI tool at all.

glyph ("Glyph") wrote:

@aredridel In the prompt under discussion here, "generated with claude code" is included in the list of things not to include, which is not an unreleased product name.

aredridel@kolektiva.social ("Mx. Aria Stewart") wrote:

@glyph So what's going on in Claude (which fwiw I do not use) is a lot of "don't expose unreleased product info”

Not _great_ mind you but that's a lot of the context for what's going on there.

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

has anybody heard from @davidgerard today? i'm really worried he might have laughed his ass off on the floor rolling to death.

aredridel@kolektiva.social ("Mx. Aria Stewart") wrote:

@glyph I measure, if informally, how often we have problems, and we talk about mitigations.

You can, in fact, check.

glyph ("Glyph") wrote:

@aredridel Treating LLMs differently here is not a double standard, it's just a standard. They're new, they're different, but most of all, if labeling weren't a big deal *why try to hide it in the first place*?

glyph ("Glyph") wrote:

@aredridel

> Are you measuring your guardrails?

Of course not. Nobody is. The resources do not exist in the software industry, let alone in volunteer open source, to do this adequately. Which is why we rely on good faith.

> do you require any unsafe practice to be labeled? Or just LLMs?

Just LLMs. First, because LLMs are novel and unique.

Second, here we're not even talking about a labeling *requirement* yet, we're talking about *active deception*.

Boosted by jwz:
fasterandworse@hci.social ("Stephen Farrugia") wrote:

@misty also note that Apple have dark-patterned this update below the fold that doesn't appear to be there

Attachments:

• video: video screen recording of ios software update screen showing a full height update for IOS 26 and a non-obvious extra scroll reveals an "also available" update for the ios 18.7.7 security update

aredridel@kolektiva.social ("Mx. Aria Stewart") wrote:

@glyph Right. Are you measuring your guardrails?

And: do you require any unsafe practice to be labeled? Or just LLMs?

That's the thing. My fundamental argument here is that _these are tools_. Sometimes that's relevant, sometimes that's not.