Back in 2018, Dario Amodei worked at OpenAI. And looking at one of its first A.I. models, he wondered: What would happen as you fed an artificial intelligence more and more data? He and his colleagues decided to study it, and they found that the A.I. didn’t just get better with more data; it got better exponentially.
@ianRobinson Common sense would suggest that while the growth may be exponential for a bit, there would be an asymptote.
Also and unfairly, because I will not listen to the program, what were the metrics used to evaluate “better?” Were they suggesting 2n source data -> n^2 correctness? As I believe this is essentially applied probability theory, were they asserting the unsurprising result that larger sample sets increases confidence values?
@pomCountyIrregs You should listen to it. It’s already a summary of the field. Any summary of this summary would require a significant bit of work from me. And it wouldn’t be better than getting the info directly from the source.
Add comment