My take on AI and FB
Why AI can never solve social and why it’s too late not to use it
I’m an AI engineer, I’ve implemented various algorithms that work on Twitter and Raw datasets in Python.
In my opinion using AI to monitor-detect-interfere with our social life is technically crazy!
Today there are debates in which algorithms to use in solving very simple problems. Not to mention each algo has its own tradeoffs… Imagine the consequences if people started driving in crowded places who can’t drive properly even on straight roads!
This is what companies are doing to REAL people, experimenting on their vulnerabilities of susceptibility without consent:
youtube.com/watch?v=iFTWM7HV2UI
This game they’re playing has psychological ramifications!
In their defence, at this point, it’s crazy not to use AI, as there’s too much analysis required in the pool of 2 billion people, and we certainly can’t do that without extensive testing. This is next to impossible to have such software model ready within reasonable amount of time without AI.
Concepts like Terrorism, Brainwashing, Extremism, Fake news, Pseudo-science conspiracy theories, etc. are all too complex even for our judicial system, and we need real judges with years of knowledge to determine the nature of the case. To detect such intricate perplexing classification head-on by a single un-changing model becomes even harder as humans improvise their strategies to bypass surveillance and detection.
We truly live in times when we haven’t yet figured out ‘How to use the Internet’
