Sander Schulhoff

speaker

947 appearances 1 recordings 1 series first heard Jun 2025 last heard Jun 2025

Sander Schulhoff’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months

No recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.

Appearances

newest first · ▶ plays the moment

Lenny's Podcast: Product | Career | Growth · AI prompt engineering in 2025: What works and what doesn’t | Sander Schulhoff (Learn Prompting, HackAPrompt) · 19 Jun 2025 podcast

Uh and we've seen historically a lot of folks saying, Oh, you know, this will be solved in a couple of years, similarly to prompt engineering, uh, actually.

Uh but very notably recently Sam Altman uh at a private event, uh, although this is that this went public information, uh said that 90 they he thought they could get to 95 to 99% uh you know security against prompt injections.

Yeah, it's it's not solvable.

It's mitigatable.

Uh you can kind of sometimes detect and track when it's happening, but it's really, really not solvable.

Uh and that's one of the things that makes it so different from classical security.

Uh I I like to say you can patch a bug, but you can't patch a brain.

Uh and you know the the explanation for that is like in classical cybersecurity, if if you find a bug.

You can just go fix that.

Uh and then you can be certain that that exact bug uh is no longer a problem.

But with AI, you know, you could find a bug where a particular I guess like air quotes, a bug, where some particular prompt can elicit uh malicious information from the AI.

you can go and and kinda train it against that, but you can never be certain.

With any

strong degree of accuracy that it won't happen again.

that's the problem.

AI red teaming, artificial social uh engineering a lot of times.

There we go.

So yeah, that is uh quite relevant.

But even getting those kind of those three you know, don't do harm to yourself, et cetera, thing is really difficult to define in some pure way in training.

Showing 721–740 of 947 · page 37 of 48