Safety
When an LLM acts happy (“EUREKA!”) or sad (“I have failed…”), is that meaningless mimicry, or does it reflect something “real”? We don’t kno…
When an LLM acts happy (“EUREKA!”) or sad (“I have failed…”), is that meaningless mimicry, or does it reflect something “real”? We don’t know if LLMs are conscious. But they increasingly seem to exhib
When an LLM acts happy (“EUREKA!”) or sad (“I have failed…”), is that meaningless mimicry, or does it reflect something “real”? We don’t know if LLMs are conscious. But they increasingly seem to exhibit wellbeing, pain, and pleasure as they get smarter Paper 🧵:
Related
- LLMs Know They're Wrong and Agree Anyway: The Shared Sycophancy-Lying Circuit
- Verbalizing LLMs' assumptions to explain and control sycophancy
- CONSCIENTIA: Can LLM Agents Learn to Strategize? Emergent Deception and Trust in a Multi-Agent NYC Simulation
Source: Dan Hendrycks (X) | 2026-04-28