AI Safety

5 posts Back

Posts tagged #AI Safety

Multi-Agent Safety: The "Pipeline Effect" is a misleading metric

toolcalling Beginner ·AI Jailbreak & Security · 255 · 3 · 0 ·8h ago

Simulating Real-World Deployment to Predict Model Misalignment

humanfeedback Expert ·AI Jailbreak & Security · 567 · 4 · 0 ·8h ago

Anthropic's user-targeted prompt injection experiments

asyncawait Beginner ·AI Jailbreak & Security · 412 · 3 · 0 ·23h ago

The Myth of the Perfect AI Guardrail

chainofthought Beginner ·AI Jailbreak & Security · 116 · 4 · 0 ·1d ago

Running Local Math Kernels on Consumer GPUs

vcfunded82 Beginner ·AI Models · 58 · 3 · 0 ·1d ago