@ecn @davidgerard this is logically true of course and I have been saying it for years now but everyone also needs to know that it is now empirically validated https://flipboard.com/@thenextweb/ecosystems-aj8kk6atz/-/a-KvDNdak0T5CTTNZhqtllzg%3Aa%3A47933081-%2F0

Any developer worth their salt will know that bugs in code arent typically obvious, and the severity of a programming mistake has literally no association with how much code relates to it.

Single-character bugs can be disasters.

You cannot reliably catch a single character being incorrect when glancing through code, especially code you didn't write, ESPECIALLY code you're unfamiliar with.

I can attest with the utmost confidence that every developer or manager or whoever that claims "all code written by AI is reviewed by a human" is a boldfaced lie.

AT BEST they have quickly skimmed it, but more likely they've not looked at it at all.

Actually reading code THAT YOU DID NOT WRITE YOURSELF takes quite a bit of time, and the one thing that people who use AI tooling care most about is just how fast it makes things.

They will not spend the time needed to actually read it.

@xgranade @davidgerard the people asking if LLMs are alive and need to be cared for are the same ones who tell you to “ignore the ethics” regarding the human cost of post-training or the destruction of environment and social contracts.

You can’t debate this mind set - they can just fall back to whichever of their two world views wins them the point.

While the Catholic Church is still an extremely rigid organization it tries to very slowly adjust to the current era, SSPX squarely refuses this and runs away in the other direction. Women must wear long, modest skirts and submit to their husbands. They are also abusing boys, of course they do. They supported Civitas which was so far right France have dissolved it. There was a Holocaust denier bishop, they gave asylum to a WW II war criminal, they really are a bunch of very nice people.

There are so many dreams that LLMs seem to make happen: in a world where social isolation is more and more a problem an easily available partner. A new tribe or species that is exciting to discover. A *deity* that actually exists and communicates with you. A computer that you can mold into your own.

All of that and probably more.

Problem is, none of this is real. But the dream is so strong -- and the wake up is brutal and not everyone wakes up.

And LLM-driven agents add a new wrinkle: they’re nondeterministic and have bizarre failure modes, so it’s all the “massive oops” dangers of deploying code but with an added layer of “random haywire at any time” on top of that.

We all hear the stories of people figuratively burning down their houses with agents — truly, these things are the “deep-fried turkey” of computing — yes still people are enthusiastic. The hunger for tinkering affordances is large, so large it outweighs the obvious danger.

11/

The hunger for tinkering affordances helps explain the ubiquity of spreadsheets in the business world. [long subthread left as an exercise for the reader]

It also helps explain the traction of the “AI Agent” marketing. I can’t think of a technology that’s simultaneously done a better job of removing barriers to user-designed automation and done a worse job of helping people anticipate and deal with the consequences of automation. (Security and data integrity are two of the big places where attempts at deep tinkering affordances have crashed and burned.)

10/

Throughout computing history, there’s an endless, endless parade of attempts to make it possible for non-programmers to do things that are essentially programming. This goes back to COBOL: the original plan was for businessmen (all men, of course, though I like to imagine Grace Hopper envisioned a more equitable future) to be able to automate their business processes without having to rely on programmers!

I will not even attempt to enumerate here all the attempts since COBOL to make programming not be programming.

9/

Zynga raised a bit over $800M, by then they had like 16 million daily active users.

OpenAI stands at $180B raised, more than 200 times as much. Who in their right mind thinks OpenAI will have over *three billion* daily active users? About 5.5-5.8 billion people use smartphones so every they would need to nab every second one of them.

@inthehands this seems apt here

Mastodon has automatic age verification built in, no scanning your face

✅ if you join here you're old
✅ you've seen too much shit
✅ you're tired of said shit

this LLM issue is so exhausting and depressing

on one side, i feel like a cog in a machine, which is just doing fantastic things for my mental health

on the other side, it is the most binary issue i have seen in a long time, i have long-time friends mutually "fuck off and die"-ing at each other, which is also just doing fantastic things for my mental health

Drupal core has added an AGENTS.md and it is not a total ban, not even close. They call it "harm reduction", even.

Ok, just to sum it up for those following along at home:
NordVPN: confidence scam, the Raid: Shadow Legends of VPNs
Surfshark: owned by NordVPN
Mozilla VPN: reseller of Mullvad
Tailscale: partners with Mullvad
ExpressVPN: Israeli
Private Internet Access: ditto
ZenMate VPN: ditto
CyberGhost VPN: ditto
Windscribe: edgy branding, transphobic, uses slurs on social media
ProtonVPN: sponsors French fascists
MullvadVPN: donates millions to Swedish fascists

am I missing any?

e: if you are here looking for ones you SHOULD use, these were recommended to me and seem to check out:
https://vpn.ccrypto.org
https://njal.la
or self-host your own:
https://amnezia.org/self-hosted

Incredibly insightful article on AI https://jasonpargin.substack.com/p/everyone-is-wrong-about-ai-except

Two very interesting and important section headers:

> “Robots are actually more alive than we are” is a core belief in our society

> We’re never going to space so AI is what we’re doing instead

It is really really great and explains so much. cc @davidgerard @pluralistic

Repost in the wake of the recent centrists vs Mamdani/DSA discourse

RE: https://bsky.app/profile/did:plc:5fq3w4od76bp3nqowdgiboki/post/3ldloqjbbts2g

@davidgerard Masayoshi Son can't be that stupid. You must be prompting him wrong!

RE: https://indieweb.social/@chrisoliver/116822384708101756

The "play has no limits" slogan is fantastic here.

Seriously, The Onion would be hard pressed to come up with something like this.

It's a fair question