Quote from OpenAI on the livestream.
'Already, Sol has been transforming our research program. As one example, GPT-5.6-Sol autonomously post-trained GPT-5.6-Luna.'
In AI 2027, we predicted that AI would take over the world or irreversibly concentrate power.
In AI 2040: Plan A, we've laid out our positive vision for what should happen instead.
@tszzl the underlying issue, to me, is these models show no whatsoever empathy to the race whose brain only evaluate five things at a time with their limited working memory. they ended up with the kind of writing requires a lot of analytical processing.
@geoffreylitt wonder if the difference in our limited working memory compared to their *almost* unlimited working memory is the main issue.
like what is 'intuitive' to them is not to us.
@tszzl the underlying issue, to me, is these models show no whatsoever empathy to the race whose brain only evaluate five things at a time with their limited working memory. they ended up with the kind of writing requires a lot of analytical processing.
@aussieljk@benglickenhaus not from my experience on codex - i often parallel run 5.5 med fast and 5.3 spark and saw the answer (+first signal) comes first from 5.5