Skip to content
Technology Security Information
  • Home
  • News
  • Security
  • Cyber Security
  • Threats

Jailbreaking LLMs with ASCII Art

Posted on March 12, 2024

Researchers have demonstrated that putting words in ASCII art can cause LLMs—GPT-3.5, GPT-4, Gemini, Claude, and Llama2—to ignore their safety instructions.

Research paper.

Posted in Security

Post navigation

Previous: The 2024 Sophos Threat Report: Cybercrime on Main Street
Next: Russia’s Foreign Intelligence Service (SVR) alleges US is plotting to interfere in presidential election

Recent Posts

  • OpenAI Wants to get College Kids Hooked on AI
  • The ‘Fantastic Four: First Steps’ Theme Is Expectedly Excellent
  • Summer Game Fest Doesn’t Just Try Its Best, It Tries Too Hard
  • Experts found 4 billion user records online, the largest known leak of Chinese personal data from a single source
  • Musk Deletes His ‘Really Big Bomb’ Claiming Trump Appears in Epstein Files

Recent Comments

No comments to show.

Archives

  • June 2025
  • May 2025
  • April 2025
  • March 2025
  • February 2025
  • January 2025
  • December 2024
  • November 2024
  • October 2024
  • September 2024
  • August 2024
  • July 2024
  • June 2024
  • May 2024
  • April 2024
  • March 2024
  • February 2024
  • January 2024
  • December 2023
  • November 2023
  • October 2023
  • September 2023
  • August 2023
  • July 2023
  • June 2023
  • May 2023
  • April 2023
  • March 2023
  • February 2023

Categories

  • Cyber Security
  • News
  • Security
  • Threats
  • Uncategorized

Related Posts

When IT meets OT: Cybersecurity for the physical world

  • Security

While relatively rare, real-world incidents impacting operational technology highlight that…

  • rooter
  • April 10, 2025
  • 1 min read
  • 0

6 reasons why SMBs love OpenText MDR

  • Security

In today’s digital-first world, small and medium-sized businesses (SMBs) face…

  • rooter
  • January 6, 2025
  • 5 min read
  • 0

Implementing Gartner Vulnerability Management Recommendations with SanerNow

  • Security

Vulnerability management is one of the most critical cybersecurity tasks…

  • rooter
  • May 22, 2023
  • 1 min read
  • 0

On the Need for an AI Public Option

  • Security

Artificial intelligence will bring great benefits to all of humanity.…

  • rooter
  • June 14, 2023
  • 8 min read
  • 0
Copyright © 2025 Technology Security Information Theme: Translucent Blog By Adore Themes.