Show newer

@lupyuen

That's a fair point made about the test already having a bias. Got no answer for that :blobcatsweat:

Though for (2), I don't think humans are becoming less creative than machines. If you're looking at an individual or sets of individuals maybe. People have different thought processes after all.

But with a sample size large enough that shouldn't be the case anymore. And machines scrape those large sample sizes of ideas and spice it up. Humans can't do that.

@lupyuen

I know this is a 'what if' point, but GPT does web scraping and sometimes mixes things up based on what's asked.

So if people had a way to 'learn' like that, would humans catch up?
Interesting read!

@tobozo @lupyuen Chromium has already been doing this for a while, it's just not enabled by default.

It's also not hid behind some chrome://flags configuration.

"GPT-4 scored in the top 1% of test-takers for the Originality of its ideas ... we believe this marks one of the first examples of AI meeting or exceeding the human ability for Original Thinking"

theconversation.com/ai-scores-

Show older
Qoto Mastodon

QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.