harisec

10 ms·

1.
Context Bombs: stopping AI attackers in their tracks(agentic.tracebit.com)
2 points by harisec 2mo ago|0 comments
2.
by harisec 2mo ago
Trying to use it via openrouter: "The model grok-4.5 is not available in your region."
3.
by harisec 5mo ago
This: User just tricked Grok and Bankrbot to send tokens with Morse code https://www.cryptopolitan.com/user-tricked-grok-bankrbot-to-...
4.
by harisec 9mo ago
Unfortunatelly, i have the same experience.
5.
Security research in the age of AI tools(invicti.com)
2 points by harisec 10mo ago|0 comments
6.
by harisec 1y ago
$44M/year? https://www.usaspending.gov/award/CONT_AWD_70RCSJ23FR0000015...
7.
by harisec 1y ago
Yes, during training multiple checkpoints are created, you can distill from any checkpoint you want.
8.
My Thoughts on the Future of "AI"(nicholas.carlini.com)
2 points by harisec 2y ago|0 comments
9.
by harisec 2y ago
Anybody can try Grok3 on Chatbot Arena (even if you are in Europe). Select Direct Chat and select the model early-grok-3. https://lmarena.ai/
10.
Coinbase, Binance Plan to List Donald Trump's Official Trump Token(coindesk.com)
2 points by harisec 2y ago|0 comments
11.
How to jailbreak most/all LLMs using Assistant Prefill(invicti.com)
3 points by harisec 2y ago|0 comments
12.
by harisec 2y ago
It doesn’t matter much how many users Bluesky is gaining, it matters how many of them will be using Bluesky in a few months. We will see.
13.
by harisec 2y ago
Congrats, good luck with your new company! I have one question regarding your ARC Prize competition: The current leader from the leaderboard (MindsAI) seems not to be following the original intention of the competition (fine tune a model wi
14.
System prompt exposure: how AI image generators may leak sensitive instructions(invicti.com)
1 points by harisec 2y ago|1 comments
15.
by harisec 2y ago
Recraft’s image generation service could leak its internal system prompts due to its unique architecture combining Claude (an AI language model) with a diffusion model. Unlike other image generators, Recraft could perform calculations and a
16.
Analyzing WordPress hack access logs with NotebookLM(invicti.com)
1 points by harisec 2y ago|0 comments
17.
by harisec 2y ago
Actually, i think it will take less than 2 years. I've been using Aider + Claude 3.5 Sonnet almost daily for a long time and the progress is very fast. We will see.
18.
by harisec 2y ago
That's a good point and I agree with you. However, would you agree that in a few years we will need far less developers than we need right now?
19.
by harisec 2y ago
These are toys but in 2 years they will probably be full projects and 2 years later people will ask "why do i need a software developer?"
20.
by harisec 2y ago
If you want to really get depressed about the future of software developers try aider.
21.
by harisec 2y ago
I never had this problem but i guess it depends on the prompt.
22.
by harisec 2y ago
Qwen 2.5 models are better than Llama and Mistral.
23.
by harisec 2y ago
There are no "LLM-generated images". These images are generated using diffusion models. LLMs are very different from diffusion models. https://en.m.wikipedia.org/wiki/Diffusion_model
24.
by harisec 2y ago
I asked a few “hard” questions and compared o1 with claude. https://github.com/harisec/o1-vs-claude
25.
Volkswagen considers historic German plant closures in cost-cutting drive(reuters.com)
1 points by harisec 2y ago|0 comments
26.
by harisec 2y ago
If you use DeepSeek Coder V2 0724 (that is #2 after Claude 3.5 Sonnet on the Aider leaderboard), the costs are very, very small. https://aider.chat/2024/07/25/new-models.html
27.
by harisec 2y ago
aider is great, i also use it almost daily. thanks for writing it Paul!
28.
by harisec 2y ago
I agree, I suspect the HuggingFace dataset that I've used is not that randomly distributed and it mostly contains prompts related with those themes. How it works is that I randomly select 5 random prompts from the dataset and use those
29.
by harisec 2y ago
All images in the LLM Dreams gallery are generated using the Flux model from the Black Forest Team, powered by the fal.ai API. The prompts for these images were brainstormed by DeepSeek, drawing inspiration from existing Stable Diffusion pr
30.
LLM Dreams – images from automatically generated LLM prompts(harisec.github.io)
3 points by harisec 2y ago|3 comments

More ›