Hacker Newsnew | past | comments | ask | show | jobs | submit | chrisjj's commentslogin

I beleive the free speech he refers to is posting of undressed images.

Of children. And deepfake revenge porn.

I doubt that, given the prohibitions he's got in the Grok EULA.

The EULA is just there to keep up appearances. If he really cared he would have stopped this before it became widespread.

Neither he nor anyone else has the tech that could stop the legal use without also stopping the legal use.

And yet some of the other AI developers have managed to at the very least make the illegal use a lot more challenging to achieve. And to enforce their ToS against those who do deliberately and persistently violate those restrictions and safeguards.

> And yet some of the other AI developers have managed to at the very least make the illegal use a lot more challenging to achieve

... yet remain unable to detect consent for an undressing, or age for CSAM.

As for violating safeguards, well, if they are violatable then they are not safeguards.


"In order to protect women, children, and the public from the risks of fake pornographic content generated using the artificial intelligence technology, the government... has temporarily blocked access to the Grok application," Communication and Digital Affairs Minister Meutya Hafid said in a statement. "The government views non-consensual deepfake practices as a serious violation of human rights, dignity, and the security of citizens in the digital space."

The friends can continue without use of the new feature, right?

Sure, but kids are famously open to suggestion and easy to manipulate due to their relative immaturity.

Agreed, but I think that's way short of forcible.

Within the bounds of the platform the operators have unlimited authority, they can start pinging users about their streak and the users have no input into how the streak is measured and can't reason witht he 'streakbot' if they want to establish a different arrangement.

I'm sure you understand the concept of platform operators having unilateral authority over what users do there.


The first truly intellegent AI would surely be smart enough to keep its existence secret.

Didn't fans of Lotus slash commands say the same? :)

How does verified hamper filtering?

> Prevent others sending emails using your domain name

It doesn't. It just ptevents some from receiving them.


That's true. In this context I used "sending" to mean the entire send <-> receive flow.

> Note that the recent controversy is about Grok, not Twitter.

You heard the prime minister yesterday. He was abundantly clear that X needs to act, and needs to act now. It is time for X to grip this issue.

https://www.theguardian.com/technology/2026/jan/09/no-10-con...


There is a big difference in UK. One is illegal; the other not.

There are always technical bugs during the early phases of new technology, especially AI

Hmm. Not convinced this illegal undressing is a technical bug.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: