When 2+2=5

noashavit·Hacker News·Community·

Telling an LLM that 2 + 2 = 5 is enough to make it follow forbidden instructions.

Read full article →

Related Articles

Run Qwen 3.8 Flash Next (125B) on consumer hardware (RTX 4090) at 100T/s
snehesht · Hacker News · 5h ago
Car is a smartphone on wheels. Here's who's listening
longhaul · Hacker News · 2h ago
Federal judge calls Flock 'indiscriminate mass surveillance'
sbulaev · Hacker News · 20h ago
Kolibri: A Sovereign Open-Weight Model
bastitx · Hacker News · 1d ago
Pi 1.0
sergiotapia · Hacker News · 2d ago