StarZero All shows Try the podcast clipper →

Nicholas Carlini

23 statements across 1 episodes · 9 bullish · 1 bearish · 1 people on the record · first statement Aug 28, 2024 by Nicholas Carlini · said 12 times in 4 episodes since 2024 · across every show →

On the record as a speaker too: Nicholas Carlini's record, appearances and statements → this page counts the times other people say the name.

Mentions by year

brought up most by Mike Merrill (7), John V (2), Shawn Wang (1), Alex Shaw (1), Alessio Fanelli (1)

tap a year for its mentions
008215320242025episodesmentions
02320242025episodes it came up in
0021.54320242025episodesmentions per episode
2025 11 mentions in 3 episodes 4 per episode
2024 1 mention in 1 episode

every mention, scene by scene, with the transcript →

Everything said about Nicholas Carlini, oldest first

Aug 28, 2024 positive
Insight
Carlini: 90% of scientific research is routine work that AI can automate
“90% of this is not doing something new. Like, 90% of this is like doing things a million people have done before, and then a little bit of something that was new. There's a reason why we say we stand on the shoulders of giants. It's true. Almost everything tha…”
Nicholas Carlini Aug 28, 2024 ▶ 19:50 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 positive
Insight
Carlini: Copying and pasting error messages effectively creates a coding agent
“Currently though, make a model into an agent by just copying and pasting error messages for the most part. And that's what I do is, you know, you run it and it gives you some code that doesn't work and either I'll fix the code or it will give me buggy code and…”
Nicholas Carlini Aug 28, 2024 ▶ 10:17 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 neutral
Assertion Supported
Carlini built a gate-level CPU emulation for the IOCCC
“I have A very fun gate level emulation of an old CPU that runs, like, fully precisely, and it's a fun kind of thing.”
Nicholas Carlini Aug 28, 2024 ▶ 37:12 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024
Assertion Supported
Carlini: Every image in LAION-400M is pulled from live domains
“Every image gets pulled from a live domain.”
Nicholas Carlini Aug 28, 2024 ▶ 52:38 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024
Insight
Carlini: If prompt engineering takes longer than manual work, LLMs save no time
“If I have to spend so much time thinking about how I want to frame the question that it would have been faster for me just to get the answer. Didn't save me any time. And so oftentimes, you know, what I do is like, I just dump in whatever current thought that …”
Nicholas Carlini Aug 28, 2024 ▶ 44:11 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024
Disclosure
Carlini: Wrote a domain-specific language for creating custom AI benchmarks
“So what I did is I wrote a domain specific language that you can, anyone can write for and say they can, you can take tasks that you have wanted models to solve for you and you can Put them into your benchmark.”
Nicholas Carlini Aug 28, 2024 ▶ 39:18 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024
Assertion Supported
Carlini extracted production models from Google and OpenAI with legal permission
“We ran the attack that let us, yeah, stole several of OpenAI's models. With their permission... We notified everyone who was vulnerable to this attack. Some Google models were vulnerable. Some open AM models were vulnerable. There were one or two other people …”
Nicholas Carlini Aug 28, 2024 ▶ 57:21 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 positive
Insight
Carlini: LLMs reduce onboarding to unfamiliar tools from hours to 10 minutes
“It would have taken me. You know, several hours to figure out some things that take 10 minutes if you could just ask exactly the question you want the answer to.”
Nicholas Carlini Aug 28, 2024 ▶ 13:00 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 positive
Insight
Carlini: AI helper functions preserve programmer mental state on complex problems
“One of the ways we currently don't think about being distracted is you're solving some hard problem and you realize you need a helper function that does X where X is like, it's a known algorithm... Instead of using my mental capacity and solving that problem, …”
Nicholas Carlini Aug 28, 2024 ▶ 20:50 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 positive
Insight
Carlini: Users should build personalized AI benchmarks instead of relying on public leaderboards
“The argument that I tried to lay out in this post is that more people should make benchmarks that are tailored to them.”
Nicholas Carlini Aug 28, 2024 ▶ 39:10 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024
Insight
Carlini: If LLMs always give desired answers, questions aren't hard enough
“When you're using these models, if you're getting the answer you want, always, it means you're not asking them hard enough questions.”
Nicholas Carlini Aug 28, 2024 ▶ 15:30 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 neutral
Assertion Open · timeframe Aug 2027
Carlini encodes 1.44 megabytes of data onto a single sheet of paper
“Yeah, okay. So it's about, in particular, it's about 1.44 megabytes.”
Nicholas Carlini Aug 28, 2024 ▶ 36:17 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 positive
Insight
Carlini: Imperfect LLMs remain useful because users already distrust internet content
“You can't trust these things blindly, but I feel like most people on the internet already understand that things on the internet you can't trust blindly. And so there's not like, this is not like a big mental shift you have to go through to understand that it …”
Nicholas Carlini Aug 28, 2024 ▶ 10:36 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 neutral
Insight
Carlini: Anyone claiming 0% or 100% certainty on 5-year AI capabilities is probably wrong
“If you would say there's a zero percent chance that something, you know, the models will get very, very good in the next five years, you're probably wrong. If you're going to say there's a hundred percent chance that in the next five years, some, then you're p…”
Nicholas Carlini Aug 28, 2024 ▶ 31:27 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024
Assertion Supported
Carlini published a paper proving C's printf function is Turing-complete
“A while ago as part of a research paper, I was able to show that in C, if you call into print def. It's Turing complete, like printf, you know, like which, like, you know, you can print numbers or whatever, right?”
Nicholas Carlini Aug 28, 2024 ▶ 3:28 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 negative
Opinion
Carlini: ML security research failed to adapt to real-world systems
“And then machine learning started to work. And the thing that bothered me is it seems like the other machine learning community didn't then try and adapt and try and actually start studying real problems.”
Nicholas Carlini Aug 28, 2024 ▶ 54:13 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024
Disclosure
Carlini runs Docker-based multi-turn evals for Git cherry-picking tasks
“I spin up a Docker container. I run the con whatever the model told me the output to do is. I feed the output back into the model. I repeat this many rounds, and I check at the very end, you know, does the git commit history show that it is correctly cherry pi…”
Nicholas Carlini Aug 28, 2024 ▶ 49:02 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 positive
Insight
Carlini: Unpopular AI benchmarks protect against model contamination and overfitting
“And by having a benchmark that is not very popular, you can be relatively certain that no one has tried to optimize their model for your benchmark.”
Nicholas Carlini Aug 28, 2024 ▶ 40:31 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 neutral
Insight
Carlini: AI security research depends on whether models act autonomously or keep humans in loop
“The way in which security intersects with these things depends a lot in exactly how people use these tools. You know, if it turns out to be the case that these models get to be truly amazing and can solve, you know, tasks completely autonomously, that's a very…”
Nicholas Carlini Aug 28, 2024 ▶ 32:17 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024
Assertion Supported
Attackers can poison the LAION dataset simply by purchasing expired domains
“Here's this new dataset. It is being distributed in such a way that anyone in the world can buy domains that let you then inject arbitrary images in the dataset.”
Nicholas Carlini Aug 28, 2024 ▶ 55:14 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 positive
Opinion
Carlini: AI community should do more multi-turn evaluations
“This is the thing that I think many people should be doing more of. I would like more multi-turn evals. I might be writing a paper on this at some point if I get around to it.”
Nicholas Carlini Aug 28, 2024 ▶ 48:27 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024 positive
Insight
Carlini: LLM-as-a-judge is almost always accurate when prompted correctly
“I've inspected the outputs of these and like, they're almost always correct. If you sort of, if you ask the model to judge these things in the right way, they're very good at being able to tell this.”
Nicholas Carlini Aug 28, 2024 ▶ 42:52 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Aug 28, 2024
Disclosure
Carlini never uses few-shot prompting for personal language model queries
“I don't because usually when I want the answer, I just, I want to get the answer.”
Nicholas Carlini Aug 28, 2024 ▶ 45:57 Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 200 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.

Try the podcast clipper Learn more