AI & ML
Estimating worst case frontier risks of open weight LLMs
OpenAI Blog0 views
In this paper, we study the worst-case frontier risks of releasing gpt-oss. We introduce malicious fine-tuning (MFT), where we attempt to elicit maximum capabilities by fine-tuning gpt-oss to be as capable as possible in two domains: biology and cybersecurity.
Source:https://openai.com/index/estimating-worst-case-frontier-risks-of-open-weight-llmsRelated posts
Your Supplier Demanded $208 More After You Paid. Do Not Click Cancel.
0dev.to
A correct webhook handler that still caused duplicate emails
0dev.to
Nasdaq erases all 2026 gains
0Hacker News
AI may have solved the most important problem in mathematics
0Hacker News
Has anyone tried combining Next-Concept Modeling (NCM) with Apple's Neural Engine (ANE)? Curious if anyone has tested this or encountered limitations. #machinelearning #apple #ai #discuss
0dev.to AI
Hydra · Coeziune OM–IA — the complete DOI registry (2026-11-20): 87 published works
0dev.to AI