rendanhogan/loophole ·
github.com · 1,635 words · saved by 1 readers
No description, website, or topics provided.
Adversarial moral-legal code system — an AI tool that stress-tests ethical principles, system prompts, and legal texts by trying to break them. The Idea Real legal systems evolve slowly. A law gets written, someone finds a loophole, a court patches it, someone finds another loophole. This process takes decades. Loophole compresses it into minutes. You state your moral principles in plain language. An AI legislator drafts a formal legal code from them. Then two adversarial agents attack it: The Loophole Finder searches for scenarios that are technically legal under your code but morally…
saved by
related reading
- GitHub - brexhq/prompt-engineering: Tips and tricks for working with Large Language Models like OpenAI's GPT-4.github.com
- Constitutional AI: Harmlessness from AI Feedbackarxiv.org
- Coding Models Are Doing Too Much | whnrehiew.github.io
- GitHub - x1xhlol/system-prompts-and-models-of-ai-tools: FULL Augment Code, Claude Code, Cluely, CodeBuddy, Comet, Cursor, Devin AI, Junie, Kiro, Leap.new, Lovable, Manus, NotionAI, Orchids.app, Perplexity, Poke, Qoder, Replit, Same.dev, Trae, Traycer AI, VSCode Agent, Warp.dev, Windsurf, Xcode, Z.ai Code, Dia & v0. (And other Open Sourced) System Prompts, Internal Tools & AI Modelsgithub.com
- GitHub - multica-ai/andrej-karpathy-skills: A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls. · GitHubgithub.com
- Designing agentic loopssimonwillison.net
- Ampampcode.com
- GitHub - salesforce/AuditNLG: AuditNLG: Auditing Generative AI Language Modeling for Trustworthinessgithub.com
- Killing Coding Agent Slop With Adversarial Self-Playusetelos.ai
- Constitutional AI: RLHF On Steroidsastralcodexten.substack.com
- GitHub - Egonex-AI/Understand-Anything: Graphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.github.com
- GitHub - requie/AI-Red-Teaming-Guide: A comprehensive guide to adversarial testing and security evaluation of AI systems, helping organizations identify vulnerabilities before attackers exploit them.github.com