LLM

“The large language models, I'm sure you'd agree there'd be four or five people who have the capital to create those large language models, and, you know, there'll be maybe one open source, there'll be three or four proprietary”

“My intuition is that with the advent of large language models and ChatGPT being that inflection point with large language models, talking to software is perhaps the most ergonomic way we can interact with software because you don't need an instruction manual t…”

“And we never let the LLM generate an answer outside of that context, for example. It's always very specific to what's in one of our videos as an, as part of the answer corpus that the LLM will use to answer a question.”

“And so, we have historically focused on primarily the revision phase of this life cycle. Not because we wanted to, but because the technology didn't really enable us to kind of help people along the full spectrum of this communication journey. Now with large l…”

“Yes, you know, this algorithm's a little better. Yes, this algorithm's a little worse. Yes, this data set's a little better. Yes, this data set's a little worse, but it's the scale that is primarily driving. And so it drove it in the large language models, hen…”

“The decisions that they make could be people in a boardroom saying, do we go with this LLM or do we go to this one? Do we know, do we scrape Reddit for how we're going to train our models, for instance, and scrape X? Or are we going to scrape something else? T…”

“And I talked to very smart people who haven't seen these models be trained and haven't seen the objectives these models are trained on. That just automatically assume that because they've seen this sliver of intelligence from these models, that they therefore …”

“And I think that it's going to take us another two or three years to really iron out these weaknesses in the models.”