I don’t think I suggested it wasn’t worrisome, just that it’s expected.
If you think about it, AI is tuned using RLHF, or Reinforcement Learning from Human Feedback. That means the only thing AI is optimizing for is “convincingness”. It doesn’t optimize for intelligence, anything seems like intelligence is literally just a side effect as it forever marches onward towards becoming convincing to humans.
“Hey, I’ve seen this one before!” You might say. Indeed, this is exactly what happened to social media. They optimized for “engagement”, not truth, and now it’s eroding the minds of lots of people everywhere. AI will do the same thing if run by corporations in search of profits.
Left unchecked, it’s entirely possible that AI will become the most addictive, seductive technology in history.
Turns out it doesn’t really matter what the medium is, people will abuse it if they don’t have a stable mental foundation. I’m not shocked at all that a person who would believe a flat earth shitpost would also believe AI hallucinations.
People who open their mind with drugs, and immediately fill that open mind with alt right social media echo chambers just get more set in their ways than before.
Musk’s always been a dick but he’s grown much worse on ketamine+Twitter.
As long as you can keep the vibe coded pieces tiny and modular you’d probably be fine. But that takes a robust knowledge of Unity and gamedev architecture, and at that point you should probably just write it yourself.
Complex, math-heavy stuff like gaming usually is too much for an AI. It’s better at like, basic Python scripts or writing a bunch of dirty CSS.
I like their software philosophy.
I think being “uncreative” like this is a smart move for long-lasting software.