LLM Security: When Being Too Safe Can Backfire
Exploring how LLM alignment affects software security, this piece dives into the balance between safety and performance, highlighting intriguing findings from recent evaluations.
Read the full articleYou might also wanna read
Security Risks of Malicious Backdoors in Large Language Models
LLM security is a critical risk for open-weight models. Learn how malicious backdoors are easily fine-tuned into AI agents to execute harmfu
pub.aimind.so·11mo ago
Reflections on LLMs and Their Impact on Software Development Practices
a short post
Research Proposal: Measuring LLM Perplexity Scaling Laws Across Codebase Sizes for Safer Software
Research proposal for measuring how coding LLM perplexity scales with codebase context size, using Lean as a test case for whether formal la
gwern.net·4d ago
Evaluating the Impact of LLM-Generated Code in Software Development
I shouldn’t have to care about this. I don’t want to care about how someone’s code gets into the IDE. Whether you wrote it by hand, copied i
alexkondov.com·11mo agoOWASP Releases Top LLM Security Risks
The OWASP Top Ten for Large Language Models (LLV) security risks affect both users and developers of LLM-AI systems such as ChatGPT, Google'
BPE Tokenization Creates Exploitable Safety Gaps in LLM Alignment, Study Finds
Character-level perturbations bypass safety alignment in modern LLMs despite leaving prompts human-readable. We identify and test a central
arxiv.org·7d ago
Comments
Sign in to join the conversation.
No comments yet. Be the first.