Data Labeling
topic on 6 shows · 6 statements across 6 episodes
More or Less 
Lenny's Podcast 
My First Million 
the Neon Show 
the MAD Podcast 
20VC
6 statements about Data Labeling, every show
Krishnan: AI data headroom lies in multi-day, long-horizon tasks
“Generally speaking, there's less of a room for simpler tasks. The more you're doing expert level tasks the more room might be there. The more you are into some you get closer and closer to these long horizon tasks. The kind of thing that would take a human one…”
Pineau: AI data market is shifting from labeling to crafting environments
“I think for me, the even bigger trend we're seeing is the moving from just labeling data to crafting environments to produce new, new tasks.”
Autonomous vehicles require less data labeling than they did historically
“Autonomous vehicles do not need as much data labeling as they did in the past.”
Puri: Human data labeling will become obsolete in 7-10 years
“Like, you know, this might be a business that I think in, like, seven to 10 years, you may not need this anymore. Like, it seems like the way AI is going, you may not need this kind of human in the loop to label all this data.”
Morin: Data labeling has literally nothing to do with building AGI
“I mean, I get that it's like data quality in the age of LLMs is very valuable. But again, this has nothing to do with building AGI, like literally nothing.”