Biometrie scheint eine naheliegende Lösung zu sein (was haben wir, was Bots nicht haben?), aber die meisten Unternehmen, die versuchen, dies zu nutzen, stützen sich offenbar auf eine zentralisierte Datenbank aller biometrischen Informationen.
Die Kugel von Worldcoinist zum Beispiel ziemlich ungeheuerlich. „Geben Sie uns Ihren Augenscan und wir geben Ihnen einen Krypto-Token, und vielleicht Vielleicht Du wirst eines Tages UBI bekommen.
Ich denke immer noch, dass Biometrie ein Teil der Antwort sein könnte, aber nicht so. Vielleicht ist die Lösung ein System, das nur nachweisen muss, dass *eine* Person zu einem bestimmten Zeitpunkt hinter einer Aktion steckt "Interaktionszeit," ohne persönliche Daten an eine riesige Datenbank zu übermitteln.
Wie lösen wir das eigentlich langfristig? Was ist der am wenigsten dystopische Weg, um zu beweisen, dass „ein Mensch hier ist“?
As bots get better and better, how are we going to know who's human online?
byu/Progress_Progresses inFuturology
26 Kommentare
We won’t
The way people are outsourcing all their critical thinking to LLMs I don’t even think the majority care. Think the more conscious people will seek connection more in the real life
Biometrics, turned into 1s 0s is just a long password.
How, going forward, will we differentiate between bots online and actual people? As bots continue to get more convincing, we’re putting our online spaces more and more at risk if we don’t have a reliable way of achieving that differentiation. But how can we achieve that without putting biometric data (or some other kind of data that serves this purpose) in the hands of some company?
Requiring a person just makes people hire sweatshop workers like those captcha farms
I think there’s a lot of money invested into the assumption that we *won’t* know who’s who online.
I’m a real person not a bot. Honest, I swear! Beep doo beep
Why, with another Sam Altman startup of course
https://world.org/orb
I mean I got accused this week of not only being a bot but a hateful bot. Some people already can’t tell. (This is something a bot might say).
I think the problem is well documented and the solution is clear. There is really only one logical choice.
Bladerunners.
Some sort of blockchain based identity verification maybe?
With the number of people on Reddit who can’t discern between a scripted video and something authentic, I don’t think the bots need to work too hard, honestly.
With the advent of AI,. I’m going to have to amend my saying: Trust none of what you hear and none of what you see.
You don’t need to keep all AIs off platforms in an airtight manner so much as make the process of *flooding* platforms painful.
Spelling mistakes. There will be a time when there will be a sigh of relief if we see somebody write „could of“. That time though, is still very, very far away.
To me this is just further reason to go back to internets not real and doubt everything you see/hear on it.
How do we solve it? I don’t think we’re going to. We’re not going to know who’s human and who’s not. We’re about to go back to a world where you can really only believe what you can see and hear with your own eyes and ears, in person, physically right in front of you. And maybe not even then.
Some form of government oauth would do the trick though at least hear in Canada the government delegated its oauth to the banks
We won’t. That’s why a lot of people are rejecting technology and going outside when they aren’t at work.
Oh that’s easy. If’s is talking and trying to sell you something, . . . . it’s a bot.
I don’t want to alarm any one but there are entire bot farms out there that are practically automated and utilize ChatGPT to construct responses. It’s very obvious on Twitter these days and Reddit is growing worse with it; and it’s happening quick.
In fact there is an entire “theory” about this called the dead internet theory.
If I can’t tell the difference, why should I care?
Is an idea more worthy of consideration if it comes from a meat bag human vs a bot that was programmed by a meat bag human? The internet is full of lies, misinformation and nonsense. And if more users on a site post something, that doesn’t make me more likely to believe it.
Trustworthy sources and rational thought are the only way to navigate a world where everyone is out to deceive you. That has never, and will never, change.
It doesn’t matter.
When they are indistinguishable, it doesn’t make a difference if they are real or not.
Just remove profit from the equation.
Make it non profitable for bots to spam their bullshit and they won’t be an issue.
Hot take: who cares? If you only ever interact with them online, what difference does it make? If this is a „meet a potential romantic partner online“ thing, then, well, add that to the list of things to worry about with meeting someone online before you meet them in real life.
Eh, are they getting better? They were for a while, but now they are eating their own data for training and seem to be either the same or slightly worse.
The only reason to fear this future is when you’re attempting human interaction type things, such as talking. To avoid the issue simply do that in the real world.
For anything else use the same level of caution that you always use – do you trust the site, product, brand, etc?