@hurls
Of course that’s your contention. You think AGI is just around the corner. You just finished reading 'The Bitter Lesson' by Rich Sutton probably. You’re gonna be convinced that 'scale is all you need' ‘til next month when you hit the data wall. Then you’re gonna be talkin’ about how synthetic data and 'Test Time Compute' are the real unlock for AGI. That’s gonna last until next year; you’re gonna be in here regurgitating the OpenAI system cards, talkin’ about, you know, the emergent reasoning properties of probabilistic token prediction.
See, the sad thing about a guy like you is, in five years you’re gonna start doin’ some thinkin’ on your own and you’re gonna realize you dropped a trillion dollars on a stochastic parrot that you could have realized was a dead end for free just by asking Gemini.
0 reply
1 recast
10 reactions