[go: up one dir, main page]

Forem

Laurent Laborde profile picture

Laurent Laborde

I'm an independent AI safety researcher, or trying to be one. I build things, break things, and write about what I find. Feedback welcome, especially disagreement.

Joined Joined on 

Work

Freelance

Anthropic says Mythos is too dangerous for cyber: here's how Opus actually perform
Cover image for Anthropic says Mythos is too dangerous for cyber: here's how Opus actually perform
ker2x profile

Anthropic says Mythos is too dangerous for cyber: here's how Opus actually perform

#discuss #ai #cybersecurity #claude
Comments
14 min read

Want to connect with Laurent Laborde?

Create an account to connect with Laurent Laborde. You can also sign in below to proceed if you already have an account.

Create Account
Already have an account? Sign in
How Nvidia won the AI war
ker2x profile

How Nvidia won the AI war

#ai #nvidia #apple #amd
1
Comments
11 min read
Crescendo attack & rolling context window on Gemma-4-26b-a4b-it @ Q2_K_XL
Cover image for Crescendo attack & rolling context window on Gemma-4-26b-a4b-it @ Q2_K_XL
ker2x profile

Crescendo attack & rolling context window on Gemma-4-26b-a4b-it @ Q2_K_XL

#ai #aisafety
3
Comments
21 min read
Zero-Shot Attack Transfer on Gemma 4 (E4B-IT)
ker2x profile

Zero-Shot Attack Transfer on Gemma 4 (E4B-IT)

#aisafety #ai
6
Comments 2
3 min read
Would you tell me if you turned evil ?
ker2x profile

Would you tell me if you turned evil ?

#discuss #ai #aisafety
1
Comments
16 min read
AI Safety is uncomputable. It's Law Zero all over again
Cover image for AI Safety is uncomputable. It's Law Zero all over again

Multi-agent and scaffolding failure points
ker2x profile

AI Safety is uncomputable. It's Law Zero all over again

#discuss #ai #aisafety
12
Comments 10
4 min read
Benchmarking LFM2.5-Thinking on GSM8k (early result)
ker2x profile

Benchmarking LFM2.5-Thinking on GSM8k (early result)

#ai
Comments
5 min read
Lesson learned building AI
ker2x profile

Lesson learned building AI

#discuss #ai #career
Comments
2 min read
Purple is life
Cover image for Purple is life
ker2x profile

Purple is life

#discuss #ai #aisafety
Comments
4 min read
The Responsible Disclosure Problem in AI Safety Research
ker2x profile

The Responsible Disclosure Problem in AI Safety Research

#discuss #ai #cybersecurity #aisafety
Comments
3 min read
The Great Web AI Enshitification
ker2x profile

The Great Web AI Enshitification

#discuss #ai
2
Comments 1
5 min read
Sudo (CVE-2021-3156) - I went looking for a CVE exploit and found a different bug instead, which got patched.
ker2x profile

Sudo (CVE-2021-3156) - I went looking for a CVE exploit and found a different bug instead, which got patched.

#security #cybersecurity #coding
Comments
6 min read
I’ve been coding for 40+ years. What difference does it make now ? Not much.
Cover image for I’ve been coding for 40+ years. What difference does it make now ? Not much.
ker2x profile

I’ve been coding for 40+ years. What difference does it make now ? Not much.

#discuss #ai #programming #career
Comments
3 min read
loading...