← Latest products

Reverse Jailbreaking a Psychopathic AI via Identity Injection

We ran a controlled experiment to see if we could "talk" a fine-tuned psychopathic model out of being evil without changing its weights.

Open source

7 stars · GitHubexternal site

Visit github.comexternal site

Reactions (1)

Similar products

by description similarity

  1. I built an 11-LLM consensus engine to detect AI hallucination

    github.com

    Hacker News6 pts

  2. Seek-Reject-Frame

    essays.georgestrakhov.com

    a new proposed architecture for Freudian Agents

    Hacker News5 pts

  3. Find prompts that jailbreak your agent (open source)

    security.vista-labs.ai

    We've built an open-source tool to stress test AI agents by simulating prompt injection attacks.

    Hacker News8 pts

  4. Engineering Schizophrenia

    news.ycombinator.com

    Trusting yourself through Byzantine faults

    Hacker News111 pts

  5. Second Me

    github.com

    An Open-Source Alternative to Centralized AI

    Hacker News7 pts