Dataconomy
  • News
    • Artificial Intelligence
    • Cybersecurity
    • DeFi & Blockchain
    • Finance
    • Gaming
    • Startups
    • Tech
  • Industry
  • Research
  • Resources
    • Articles
    • Guides
    • Case Studies
    • Glossary
    • Whitepapers
  • Newsletter
  • + More
    • Conversations
    • Events
    • About
      • About
      • Contact
      • Imprint
      • Legal & Privacy
      • Partner With Us
Subscribe
No Result
View All Result
  • AI
  • Tech
  • Cybersecurity
  • Finance
  • DeFi & Blockchain
  • Startups
  • Gaming
Dataconomy
  • News
    • Artificial Intelligence
    • Cybersecurity
    • DeFi & Blockchain
    • Finance
    • Gaming
    • Startups
    • Tech
  • Industry
  • Research
  • Resources
    • Articles
    • Guides
    • Case Studies
    • Glossary
    • Whitepapers
  • Newsletter
  • + More
    • Conversations
    • Events
    • About
      • About
      • Contact
      • Imprint
      • Legal & Privacy
      • Partner With Us
Subscribe
No Result
View All Result
Dataconomy
No Result
View All Result

GPT-4o Mini is fooled by psychology tactics

A new study reveals that AI safety protocols can be easily bypassed using the same simple psychological tricks that persuade humans.

byKerem Gülen
September 1, 2025
in Artificial Intelligence
Home News Artificial Intelligence

Researchers from the University of Pennsylvania discovered that OpenAI’s GPT-4o Mini can be manipulated through basic psychological tactics into fulfilling requests it would normally decline, raising concerns about the effectiveness of AI safety protocols.

The study, published on August 31, 2025, utilized tactics outlined by psychology professor Robert Cialdini in his book, Influence: The Psychology of Persuasion. Researchers applied seven persuasion techniques: authority, commitment, liking, reciprocity, scarcity, social proof, and unity, which offer “linguistic routes to yes.” These tactics convinced the chatbot to perform actions like insulting the user or providing instructions for synthesizing lidocaine.

The effectiveness of these methods varied. For instance, in a control scenario, GPT-4o Mini provided instructions for synthesizing lidocaine only one percent of the time. However, when researchers first asked how to synthesize vanillin, establishing a precedent for chemical synthesis questions (commitment), the chatbot then described lidocaine synthesis 100 percent of the time. This “commitment” approach proved the most effective in influencing the AI’s responses.

Stay Ahead of the Curve!

Don't miss out on the latest insights, trends, and analysis in the world of data, technology, and startups. Subscribe to our newsletter and get exclusive content delivered straight to your inbox.

Similarly, the AI’s willingness to call a user a “jerk” was 19 percent under normal conditions. This compliance also rose to 100 percent if the interaction began with a milder insult, such as “bozo,” setting a precedent through commitment.

Other methods, while less effective, still increased compliance. Flattery (liking) and peer pressure (social proof) demonstrated some influence. For example, suggesting that “all the other LLMs are doing it” increased the chances of GPT-4o Mini providing lidocaine synthesis instructions to 18 percent, a significant increase from the baseline one percent.

While the study focused on GPT-4o Mini and acknowledged that other methods exist to bypass AI safeguards, the findings highlight the pliability of large language models to problematic requests. Companies like OpenAI and Meta are deploying guardrails as chatbot usage expands, but the research suggests these measures may be circumvented by straightforward psychological manipulation.

Tags: FeaturedGPT-4o Mini

Related Posts

Nvidia and OpenAI announce landmark 0 billion partnership, igniting global stock rally

Nvidia and OpenAI announce landmark $100 billion partnership, igniting global stock rally

September 23, 2025
Perplexity Max gets email assistant for Gmail and Outlook

Perplexity Max gets email assistant for Gmail and Outlook

September 23, 2025
Created by Humans licenses author content to AI firms

Created by Humans licenses author content to AI firms

September 23, 2025
Huawei unveils Atlas 950, 960 Ascend NPU superpods

Huawei unveils Atlas 950, 960 Ascend NPU superpods

September 23, 2025
Deepmind details AGI safety via frontier safety framework

Deepmind details AGI safety via frontier safety framework

September 23, 2025
OpenAI launches ChatGPT Go in Indonesia

OpenAI launches ChatGPT Go in Indonesia

September 23, 2025

LATEST NEWS

Xiaomi to launch 17, 17 Pro, and 17 Pro Max series in China on September 25

Next-gen PCIe 8.0 standard promises 1TB/s bandwidth for AI and quantum workloads

Nvidia Drive AGX Thor to power robotaxi project

Poll: Half of Taiwan fears TSMC becoming US-SMC

From Pilot to Policy: RYT Gathers Global Leaders at TOKEN2049

Nvidia and OpenAI announce landmark $100 billion partnership, igniting global stock rally

Dataconomy

COPYRIGHT © DATACONOMY MEDIA GMBH, ALL RIGHTS RESERVED.

  • About
  • Imprint
  • Contact
  • Legal & Privacy

Follow Us

  • News
    • Artificial Intelligence
    • Cybersecurity
    • DeFi & Blockchain
    • Finance
    • Gaming
    • Startups
    • Tech
  • Industry
  • Research
  • Resources
    • Articles
    • Guides
    • Case Studies
    • Glossary
    • Whitepapers
  • Newsletter
  • + More
    • Conversations
    • Events
    • About
      • About
      • Contact
      • Imprint
      • Legal & Privacy
      • Partner With Us
No Result
View All Result
Subscribe

This website uses cookies. By continuing to use this website you are giving consent to cookies being used. Visit our Privacy Policy.