Yeah. I've seen this happen with people doing it. It's just bad access management.

And anyone can do it with the wrong access granted at the wrong moment in time...even Sr. Devs.

At least this one won't weight on any person's conscience. The AI just shrugs it off.

The AI does nothing the like. It predicts tokens. That's it.

Describing the tech in anthropomorphic terms does not make it a person.

I feel like you didn't get the joke at the end.

Interesting project but I can't tell, is the language used supposed to be satire?

The "Arduina" comment definitely made me think it might be at least a little satirical in nature

It's a bit Sokal Hoax, in that you have to be quite familiar with this kind of thing to be able to distinguish it from satire. But a lot of "artist's statements" are written that way. If you go into a museum and find an incomprehensible objet d'art with a small explanatory plaque next to it, the plaque will be written like this.

In a way it's no different from LinkedIn or a VC pitch, the need to hit certain phrases to appeal to the audience.

It's not, the entire site appears to be a serious examination of technology and hacking ethics through a feminist lens

Honestly, the language isn't super off or abnormal in other circles, maybe it's a lot more telling that when posted on a tech-oriented site it's seen as ridiculous

I think it feels naive, to me. It's a fun project in terms of experimenting with techniques and materials, but it's very far from relevant in terms of the more abstract goals and ideas they discuss.

INAL, but if the authorities had captured your device with touchID enabled and legally ask you to use it to login and you do an action that would disable touchID, then that would be "obstruction".

That's the point of this. TouchID is no longer enabled. Someone unknown party approaches you, you close your lid (disabling TouchID). Then they "legally" ask you to put your finger on the sensor. You do. They didn't ask you before you close your lid.

You're thinking more along the lines that they ask you to touch the sensor and you use your fingernail razor blades to damage the sensor or something like that.

Yes, I meant to respond to other comments in here directly, but got messed up.

Others had floated the idea of locking by using an alternate finger with touchID, after the fact.

The major thing I like is being able to use it from telegram. Its a different usage model.

It feels like a co-worker.

Huh... I never thought about it like that. I had no interest in using OpenClaw, but my coworkers are extremely dumb and inconsistent. OpenClaw might be comparable.

Openclaw did what no major model producer would do. Release insanly insecure software that can do whatever it wants on your machine.

If openai had done it themselves, immediate backlash.

Is it? You basically got 95% of the way there with Claude Code inside of a container. People were using CC outside of development scope for awhile.

> You basically got 95% of the way there with Claude Code inside of a container.

OpenClaw and Claude Code aren't solving the same problems. OpenClaw was about having a sandbox, connecting it to a messenger channel, and letting it run wild with tools you gave it.

A messenger and ssh'ing into Claude Code from your phone aren't that much different.

I’m not an OpenClaw user but it’s obvious that OpenClaw was very different than that.

OpenClaw was about having the agent operate autonomously, including initiating its own actions and deciding what to do. Claude Code was about waiting for instructions and presenting results.

“Just SSH into Claude Code” is like the famous HN comment that didn’t understand why anyone was interested in DropBox because you could do backups with shell scripts.

You have used Claude though? Or codex or antigrav? backups seem different

The real magic is heartbeat which is essentially cron on steroids. The real difference between running Claude Code in the terminal and OpenClaw is that the agent is actually intuitive and self-driven.

People would wake up to their agent having built something cool the night before or automate their workflow without even asking for it.

Is the idea it can't handle 24hr+ tasks? I know Codex can do heartbeat but I haven't tested the limits of its heart.

That’s what CC does…I don’t need a messenger wrapper to do those things.

Major producers like OpenAI optimize for safety and brand reputation avoiding backlash. Open source projects optimize for raw capability and friction less experimentation. It is risky yes, but it allows for rapid innovation that strictly aligned models can't offer.

Life reads a lot like satire now.

Loving AI bots. Killing yourself based on what an AI bot says.

Its hard to believe any of this is real or should be.

Google also has the most to lose.

Is this for people who haven't read the docs on how to instruct the agent on common setups and preferences?

They're using odd language to justify their position. "Wage growth"?

Do they mean "real wage growth"? Because that got worse under Biden.

100%

Personal experience: In my town a public parking lot could not be built due to it possibly being "endangered moth" habitat.

There are places where you can still build things in the US, but they are more and more scarce.

Are you arguing that USA can no longer build parking lots due to environmental concerns? If so, that would indeed be remarkable since parking lots seem to be the facility that almost every US town has been able to build more than enough of.

You are the consumer, not the developer. As such, you have extreme survivorship bias.

Unless you're a developer/builder you have no concept of the projects KILLED by bureaucracy.

All you see is what was allowed to be built.

I’d like to see parking lots go extinct.