OpenToolslogo
ToolsExpertsSubmit a Tool
Advertise
  1. home
  2. news
  3. tags
  4. bon-technique

bon technique

1+ articles
AIAI safetyAI vulnerabilitiesAnthropicBoN technique

AI 'Jailbreaking': New BoN Technique Outsmarts Top Models Like GPT-4 and Claude 3.5

Researchers from Anthropic, Oxford, Stanford, and MIT introduce the Best-of-N (BoN) method—a groundbreaking ‘jailbreaking’ technique that bypasses AI safety protocols to trick models into harmful outputs. The method shows a staggering 50% success rate on models like Claude 3.5, GPT-4, and Gemini.

Dec 24
AI 'Jailbreaking': New BoN Technique Outsmarts Top Models Like GPT-4 and Claude 3.5

Related Topics

AIAI safetyAI vulnerabilitiesAnthropicBoN techniqueClaude 3.5GPT-4GeminiOxfordjailbreaking

Stay in the loop

Weekly updates on tools, models, and the companies building them.

Subscribe free

Footer

Company name

The right AI tool is out there. We'll help you find it.

LinkedInX

Knowledge Hub

  • News
  • Resources
  • Newsletter
  • Blog
  • AI Tool Reviews

Industry Hub

  • AI Companies
  • AI Tools
  • AI Models
  • MCP Servers
  • AI Tool Categories
  • Top AI Use Cases

For Builders

  • Submit a Tool
  • Experts & Agencies
  • Advertise
  • Compare Tools
  • Favourites

Legal

  • Privacy Policy
  • Terms of Service

© 2026 OpenTools - All rights reserved.

Sign in with Google