Guardrails

“Guardrails is, you know, better than prompt engineering.”

“The meta lesson for me is, you know, guardrails are technical debt. Too many guardrails are debt, right? They can break the product.”

“Prompt-based defenses are the worst of the worst defenses, and we've known this since early twenty-twenty-three. There have been various papers out on it. We studied it in many, many competitions, or we, you know, the original hack-a-prompt paper and TensorTru…”

“Putting up a guardrail is not, it's not going to do anything in terms of preventing that user from doing that, because, I mean, first of all, if the user's like, ah, guardrail, you know, too much work, they'll just go to one of these websites and get that info…”

“When it comes to AI security, the AI security industry in particular, I think we're going to see a market correction in the next Year, maybe in the next six months where companies realize that these guardrails don't work.”

“Guardrails don't work. They just don't work. They really don't work. And they're quite likely to make you overconfident in your security posture, which is which is a really big, big problem.”

“I do think they're finding clever and clever ways to lock down particular areas sometimes, but I think it's at the expense of capability and creativity. So there's some model providers that aren't prioritizing this and they seem to do better on benchmarks for …”

“So stepping in there with policy that says do not go on this, that's a big deal for things in the medical setting where you may have, you know, people use the word guardrails to talk about AI where they're like, we're going to put guardrails so it doesn't do t…”

“Path one is guardrails, right? And there's a, and VCs are going to make a lot of money on guardrails.”

“There's no like external Product focused companies are like, oh, you know, I have the best guardrail now. It's not a realistic solution. It has to be the AI labs. It has to be, I think it has to be innovations in model architectures.”

“When I talk about guardrails, the first thing that comes to mind is, oh yeah, we have to do PII redaction and sensitive data redaction. But somehow we've seen that the more common use cases for guardrails end up being, you know, is it the right length? Is it u…”

“And again, you can decrease latency massively by building this on the gateway. Rather than building this within your application.”

“In reality, we've seen what gets deployed the most is regex-based guardrails to say, you know, I want to catch for specific words, and then take certain actions based on that. Or I want to catch for empty outputs.”

“But you are able to put in place guardrails better if you yourself have used the technology.”

“And I do think that open models will immediately have their guardrails breached.”

“We see up to 40, 84 percent improvement in further accuracy because of those capability as well.”

“The reason why ChatGPT is hard to put guardrails on is because they're trying to go after every possible use case, right?”

“So we have put guardrails on the business now for 45 quarters. We've had guardrails on the business. Very, very careful and thoughtful growth, not overextending ourselves, but at the same time, careful management of sales and marketing efficiency, and then get…”