Aaron Levie
speaker
937 appearances
2 recordings
1 series
first heard Jul 2025
last heard 5 Sep
Aaron Levie’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsRecordings per month over the last 12 months — 1 in all, peaking in Sep 2026 with 1.
Appearances
real risk that we perceive.
Cause I would argue that if if if you kind of snapshot it in time, fable uh right now, the level of things that they they they push down
and prevent you from doing that would be untenable for the future of AI.
People will will simply not use AI if if this is the kind of ongoing environment.
And so so like how should we decide now?
Like it like maybe maybe the they see these risks are very high, highly nuanced.
And so so I that's why I'm I generally lean toward like it's all very like way too early to be locking these models down um you know given uh given how early we are.
Yeah, we're I'm I'm still very bullish on software engineering.
Uh we're we're what you know, the way that we've used these models is is simply to just to do way more than we were before.
Um the the kinds of and it's this is like so qualitative and only only people that are, you know, probably like in the product roadmap reviews each day can, you know, maybe kind of fully process, but like there's something totally different.
When you're when you're doing a pl a product planning uh session, uh when you are thinking in a world of of just sheer human, you know, based constraints versus versus now what AI unlocks, like it completely changes your mindset about the things that you'll go and tackle and take on.
We have multiple dozens of of uh of projects right now that we absolutely would not be doing if AI didn't
exist.
We just we would not have we would not have lit up the projects.
Like we would have said no, that that's too complex.
It's not worth it.
Like you actually have two interesting scenarios.
You either say no to the very small things because it's not worth it.
Or you say no to the really big things because they're simply too hard.
And so most of your software projects kind of stay in this middle band, which is like,
Showing 201–220 of 937 · page 11 of 47
← Previous
Next →