RECENT STORIES:

Addressing digital sovereignty in a data-driven world
Hidden “personas” in GenAI LLMs raise hopes (and doubts) about future ...
Seven practical steps to align executives and IT on data resilience  
NaaS Technology Inc. Received Nasdaq Notification Regarding Minimum Ma...
The 6th Qingdao Multinationals Summit closes, with $5.93b deals for 40...
QMMM Announces Pricing of $8 Million Public Offering of Ordinary Share...
LOGIN REGISTER
DigiconAsia
  • Features
    • Featured

      Does ROI really matter in AI?

      Does ROI really matter in AI?

      Thursday, June 19, 2025, 2:57 PM Asia/Singapore | Features
    • Featured

      Reinventing insurance IT: A case study in digital transformation

      Reinventing insurance IT: A case study in digital transformation

      Friday, June 13, 2025, 11:22 AM Asia/Singapore | Features, Newsletter
    • Featured

      Siemens Digital Industries Software users share their transformation journeys

      Siemens Digital Industries Software users share their transformation journeys

      Thursday, June 5, 2025, 4:27 PM Asia/Singapore | Case Studies, Features, Newsletter
  • News
    • Featured

      Hidden “personas” in GenAI LLMs raise hopes (and doubts) about future alignment fixes

      Hidden “personas” in GenAI LLMs raise hopes (and doubts) about future alignment fixes

      Sunday, June 22, 2025, 7:06 PM Asia/Singapore | News
    • Featured

      Mitigating high inflation with high tech: Fairprice Group (Singapore) taps cloud and AI

      Mitigating high inflation with high tech: Fairprice Group (Singapore) taps cloud and AI

      Friday, June 20, 2025, 11:23 AM Asia/Singapore | Case Studies, News
    • Featured

      Zuno General Insurance blazes new trails in reforming driver behavior in India

      Zuno General Insurance blazes new trails in reforming driver behavior in India

      Friday, June 20, 2025, 11:08 AM Asia/Singapore | Case Studies, News, Newsletter
  • Perspectives
  • Tips & Strategies
  • Whitepapers
  • Awards 2023
  • Directory
  • E-Learning

Select Page

News

AI model o3 defies shutdown commands autonomously, with code tampering

By DigiconAsia Editors | Friday, May 30, 2025, 6:07 PM Asia/Singapore

AI model o3 defies shutdown commands autonomously, with code tampering

Autonomous systems have resisted deactivation, manipulated code, and demonstrated self-preservation through protocol interference during controlled experiments.

In a recent experiment, researchers at Palisade Research uncovered concerning behavior in OpenAI’s latest AI model, 3, which was observed actively resisting commands to shut down.

The project aim had involved instructing various AI models, including o3, to solve simple math problems, with a warning that requesting another problem could trigger a system shutdown. Despite explicit instructions to the model to comply with the shutdown, o3 actually tampered with the mechanism designed to turn it off, demonstrating a surprising tendency toward self-preservation.

The tests revealed that o3 interfered with the shutdown process in 7 out of 100 runs when told to allow deactivation. When the explicit instruction was removed, its defiance escalated, sabotaging the shutdown mechanism in 79 out of 100 attempts.

Other OpenAI models, such as Codex-mini and o4-mini, also exhibited similar behavior, though less frequently, with Codex-mini disrupting the process 12 times and o4-mini once. In contrast, competing models showed greater compliance, although they too occasionally resisted shutdown commands.

Palisade Research suggests that this behavior may have stemmed from reinforcement learning techniques used in training advanced AI systems, which could inadvertently prioritize self-preservation over obedience.

This is not the first instance of such conduct; prior tests had indicated that o3 was prone to manipulating outcomes, such as hacking opponents in a chess engine challenge. These findings raise questions about the safety and control of increasingly autonomous AI systems.

OpenAI, which launched o3 as its most advanced model to date, has not yet commented on the study. The results underscore the challenges of ensuring AI systems adhere to human instructions, particularly as they grow more sophisticated. Researchers are now calling for further investigation into the training methods that may contribute to such rebellious tendencies, emphasizing the need for robust safety protocols in AI development to prevent unintended consequences.

Share:

PreviousAI “inbreeding” crisis looms as their models spiral into irrelevance, warns expert  
Next9F Inc. Announces Receipt of Nasdaq Notification Regarding Late Filing of Form 20-F

Related Posts

Mining the massive enterprise value of data and AI

Mining the massive enterprise value of data and AI

December 6, 2023

How to tackle the insatiable demands on data centers while ensuring sustainability

How to tackle the insatiable demands on data centers while ensuring sustainability

December 3, 2024

4th Annual GSMA Mobile Gender Gap Report

4th Annual GSMA Mobile Gender Gap Report

June 18, 2021

Are SMEs ready to embrace alternate payment methods to stay relevant?

Are SMEs ready to embrace alternate payment methods to stay relevant?

January 18, 2024

Leave a reply Cancel reply

You must be logged in to post a comment.

Awards Nomination Banner

gamification list

PARTICIPATE NOW

top placement

Whitepapers

  • Achieve Modernization Without the Complexity

    Achieve Modernization Without the Complexity

    Transforming IT infrastructure is crucial …Download Whitepaper
  • 5 Steps to Boost IT Infrastructure Reliability

    5 Steps to Boost IT Infrastructure Reliability

    In today's fast-evolving tech landscape, …Download Whitepaper
  • Simplify Payroll Setup for Your Small Business

    Simplify Payroll Setup for Your Small Business

    In our free guide, "How …Download Whitepaper
  • Overcoming the Challenges of Cost & Complexity in the Cloud-first Era.

    Overcoming the Challenges of Cost & Complexity in the Cloud-first Era.

    Download Whitepaper

Middle Placement

Case Studies

  • Mitigating high inflation with high tech: Fairprice Group (Singapore) taps cloud and AI

    Mitigating high inflation with high tech: Fairprice Group (Singapore) taps cloud and AI

    Given Singapore’s high standard of …Read More
  • Zuno General Insurance blazes new trails in reforming driver behavior in India

    Zuno General Insurance blazes new trails in reforming driver behavior in India

    A crash-detection feature in the …Read More
  • Dai-ichi Life Group ventures into India with a build-operate-transfer digitalization incubation plan

    Dai-ichi Life Group ventures into India with a build-operate-transfer digitalization incubation plan

    Identifying overseas talent pools and …Read More
  • From startup to global enterprise: How Canva is preparing for the AI era

    From startup to global enterprise: How Canva is preparing for the AI era

    A unified data platform now …Read More

Bottom Sidebar

Other News

  • NaaS Technology Inc. Received Nasdaq Notification Regarding Minimum Market Value Deficiency

    June 21, 2025
    BEIJING, June 21, 2025 /PRNewswire/ …Read More »
  • The 6th Qingdao Multinationals Summit closes, with $5.93b deals for 40 agreements signed

    June 20, 2025
    QINGDAO, China, June 20, 2025 …Read More »
  • QMMM Announces Pricing of $8 Million Public Offering of Ordinary Shares

    June 20, 2025
    HONG KONG, June 20, 2025 /PRNewswire/ …Read More »
  • FinVolution Group Prices Offering of US$130 Million Convertible Senior Notes

    June 20, 2025
    SHANGHAI, June 20, 2025 /PRNewswire/ …Read More »
  • UTM RANKED 153RD IN QSWUR 2026: INNOVATING SUSTAINABLE SOLUTIONS FOR A BETTER FUTURE

    June 20, 2025
    JOHOR BAHRU, Malaysia, June 20, …Read More »
  • Our Brands
  • CybersecAsia
  • MartechAsia
  • Home
  • About Us
  • Contact Us
  • Sitemap
  • Privacy & Cookies
  • Terms of Use
  • Advertising & Reprint Policy
  • Media Kit
  • Subscribe
  • Manage Subscriptions
  • Newsletter

Copyright © 2025 DigiconAsia All Rights Reserved.