AI Nightmare Scenario

[Lism]

I was watching that Geoguessr guy, Rainbolt, and his incredible ability to extract geographic information from photos. Imagine an AI-based program that has access to every voice clip, profile, photo, database, and social media post that is publicly accessible on the web. Because AI has basically infinite long-term memory and is able to do basically infinite amount of work without getting tired, it is possible someone could program an AI to predict with decent accuracy the real life identity of everyone ever to make at least a few hundred posts on the internet. It's even worse if this person has voice clips. There's nothing illegal about such an AI because it is using completely public information. Furthermore, someone could create another AI that judges every active account on Social Media for whether it has non-woke opinions, and if the opinions meet a certain threshold of non-woke, the first AI will automatically run on them and put their real identity in a database of offensive people. In countries with "hate speech laws", these people will probably be arrested and fined or imprisoned.

In an instant, the entire internet will be turned into Netwide LinkedIn, and Western politics will regress heavily.

This is a possible future scenario, unless wokeness is destroyed for good, or unless the other deleterious effects of AI (endless potential for deepfake misinformation, utility to scammers and hackers, destruction of the education system) cause AI to be banned or strictly regulated or destroyed by domestic terrorism against AI facilities by Luddites

 

[chrisgriffinirl]

Wokeness wont be destroyed as long as EU exists
Ai WILL take over
YOu take the best 2035 prediction award