There are even worse things on Nostr now:

nostr:nprofile1qqsyzdg7x9nw0dkamhzvpt2gu6f4rmp52qnhrgskm0sgh2wxs0z0u8qprpmhxue69uhkummnw3ezuendwsh8w6t69e3xj730qyghwumn8ghj7mn0wd68ytnhd9hx2tcppemhxue69uhkummn9ekx7mp0lhe3m6

Reply to this note

Please Login to reply.

Discussion

That's so dumb. AI can't even fact check itself. 😑

Its worse.

No Nostrich built that, its using a Microsoft content moderation API.

They've built is an AI Political Commissar who snitches can summon to inject AI slop into any conversation that deviates from The Narrative.

A Commissar who cannot be bribed, reasoned with, or (easily) fragged.

And someone brought this to Nostr

Sounds like someone's wet dream...

Utterly disgusting.

No, the bot don't use any Microsoft API, this is false. It use the Mistral API. The code and the AI prompt are fully open source : https://github.com/judemont/Nostr-FactChecker .

Yes, it can make mistakes. I try to improve its reliability by giving it the ability to search the web and open web pages. So it generates text based on web sources.

No, it won’t uncover a potential global mafia controlling all sources.

It’s just a tool. Don’t overestimate it. Sometimes, it’s just a faster way to find out the current scientific consensus on a theory than doing a web search yourself. It can be useful for those who choose to use it. If you really don’t want to see it, you can mute it.

It has no authority. You don’t have to agree with it. I don’t always agree with it either.

I really try to be transparent about these points because the purpose of my bot is often misunderstood. That’s why I include a cautionary message in all its replies.

Read its profile message for more info.

...and I didn't bring it to Nostr, I built it for Nostr.

Why?

(That's the important question.)

Then he needs to be kicked in the nuts every time he thinks that he should be doing anything to curb "misinformation." 😑

Yeah if you look at how he is using it and then when it's been used on him (at least by me) he's just as guilty of "spreading misinformation" as the people he sends his bot after.

So he's a hypocrite and he's obviously out for some amount of power that really doesn't belong here.

He's accused me of trying to "censor him" for telling him to take the bot down because I have a philosophical disagreement with fact checking as a whole.

That’s a good question. I have two answers.

First, because I saw a lot of misinformation on Nostr. Misinformation isn’t the same for everyone, and that’s fine, diversity of opinion is incredibly valuable. But sometimes it’s not just about opinions; it’s about demonstrably false claims designed to deceive and manipulate.

Debunking false theories with evidence and sourced responses takes 20 times longer than spreading misinformation. An AI-powered tool was an imperfect but useful solution I wanted to explore.

I built this tool to give anyone an easy way to get more context about the information they encounter.

Second, I’m a computer science student, and I think Nostr and Bitcoin are among the most interesting protocols today. I wanted to experiment by building something on top of them.

I won’t stop working on it. For one, people are using it, so it’s useful (and occasionally entertaining). But also, I refuse, as a matter of principle, to give in to the harassment and violent censorship attempts from certain individuals.

Most of this note disgusts me. If you think it's ok to build tools to efficiently eviscerate people, then you're a sadist. You might not know it, but that's how it seems to me.

But sure. You do you.

How could the bot possibly eviscerate anyone ?

It's just a tool.

If you want to use you can.

If you don't want to use it you can just ignore it.

IIf it really bothers you, you can mute it.

That's how Nostr works: you choose the content you want to see.

Fact-checking is not sadism, wanting to prevent it is.

Thats better than I thought.

Mistral is better than the Microsoft APIs I've seen before, but its far from transparent, decentralized or authoritative.

Appreciate the cautionary message, but surely we can do better

We can do better by just not doing anything of this sort. 🙄

This dude needs to be berated until he turns into a goop or paste.

You can't code

Correct. I'm utterly awful at it.

So what?

You probably can't build a race care from scratch. I can. (And have.)