Tracking AI existential risk. Source-backed context. Original reporting always linked.
MONITORING CORE AI-RISK FEEDS · UPDATED HOURLY
Analysis

There Will Be Hells

Zac Boring October 20, 2026 1 min read
Read original source →

Key takeaway

As our understanding of mechanistic interpretability and steering techniques increase, for both LLMs and uploaded connectomes, there is a corresponding "hellish capabilities" increase.

Why it's on PDOOM

PDOOM selected this story for its alignment & control signals: Interpretability, Capabilities.

InterpretabilityCapabilitiesFrom LessWrong AI

As our understanding of mechanistic interpretability and steering techniques increase, for both LLMs and uploaded connectomes, there is a corresponding "hellish capabilities" increase. We get better at torturing the digital minds we create.Our brief history with open source technologies such as MaleCNS or the LLM pain steering repo shows that torture is often one of the first use cases pursued when the public is given access to a digital mind perceived as capable of suffering or signalling suffe

By Stephen Martin

Read the full article at LessWrong AI →