RandomDistort

I get loads of them when I'm on a Netsweeper filtered network... pretty much any time any asset a page loads is from a blocked site (social media pixels normally).

I'd hazard a guess that it's based on what the LLM can "design" without actually being able to see it or have taste and it still reliably look fine to humans.

Are you sure the kids aren't learning to delete the messages?

Works until someone can afford a better and more expensive AI tool, or can afford to pay a knowledgeable human to help them answer.

When you have a simple tool you have written for yourself, that you need to be reliable and accessible but also that you don't use frequently enough that it's worth the bother of running on your own server with all of that setup and ongoing maintenance.

Is this open source?

To add to this, if you use Zoho's free plan this can cost you literally nothing.

Has someone got a guide they can link that is easy enough to understand for service administrators that aren't doing anything complicated?

Also, can I request these risk assessments off companies myself?

By the sounds of it, if there was a painless way of doing it, he would probably be fine with this.

If he consented when he was alive, I don't see how it could be moral to ignore his wishes and let more people die as a result.

I guess laws will get in the way of this kind of thing though.

There's not even a privacy notice on the website, is that legal?

Is there some document somewhere that lists all the potential ways of doing stuff like this?

AIs don't behave. They are a lot of fancy maths. Their creators can behave in ethical or moral ways though when they create these models.

= not to say that the people that work on AI are not incredibly talented, but more that it's not human

I know the law is often behind the trend because of the rate tech develops, but surely the old analogy for all technology is postal mail.

Last time I checked, the postal service had no responsibility or requirement that they don't distribute certain messages or ideas? In some cases the government can ask to intercept them, but there's no regulation requiring them to scan letters for banned content.

Why don't these same rules apply to online technology?

I doubt the cert issue is on your end, it appears whenever OpenDNS (Cisco Umbrella) tries to intercept any site when it blocks it.