GenAI solves ML's first mile, but traditional tools handle the last
Shreya Rajpal · Guardrails AI: The Playbook for Safer, Hallucination-Free LLMs — Shreya Rajpal Explains · Sep 27, 2023 · at 34:43
Shreya Rajpal is the Co-Founder and CEO of Guardrails AI. She references research from Stanford's Hazy Research Lab regarding LLM deployment challenges.
0:00 / 0:28exact quote · 28.2s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“LLMs and Generative AI really helps solve, like the first mile problem in ML, right? But the last mile problem, which is, like, how do you take this generic generalizable technology and make it work, like, specifically for your use cases and for your actual application, like you need other more traditional ML tools to make sure that you have the, you know, you have the guarantees and you have the reliability that you typically need.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Shreya Rajpal
Production LLM deployment challenges mirror autonomous vehicle development
“The trajectory of issues and concerns that people are running into are similar to, you know, like the path is similar to what it was in self-driving, which is, you know, How do I get like this, you know, runtime safety? How do I get like runtime constraints, e…”
Existing model risk frameworks fail for third-party AI models
“Existing, for example, like model risk management frameworks don't really apply when you haven't built the model yourself. You know, you didn't like curate the data that the model was trained on. And so you can't make any claims to that.”
Fine-tuning cannot eliminate LLM hallucinations
“At the end of the day, these models are like next token predictors, which is, you know, like they kind of look at like what they've predicted until now, and then, you know, figure out like what the next token they're on is. And from that, like, even with fine-…”
Fine-tuned Llama 2 achieves performance comparable to GPT-3.5 and GPT-4
“Straight out of the bat, if you just use Lama tool directly, I don't think you could get like comparable performance, you know, with GPD, 3.5 or four today. But like with fine tuning, if you make that investment in curating your data set in running that fine t…”
RAG is the definitive way to build generative AI today
“Rag is the way to build you know, these models today”
Open source turns abstract AI safety into an actionable engineering problem
“What the open source really ends up doing as you know, a framework is taking down this very abstract problem of what it means to do safe AI development, right? Like it's a very abstract problem. It's almost an academic problem to some degree, and it takes that…”