Close Menu
    What's Hot

    Nvidia Breaks $215 Billion Revenue as AI Demand Drives Unprecedented Growth

    February 26, 2026

    Aston Martin to cut 20% of jobs after losses widen to £363.9m

    February 25, 2026

    Macron Plans to Expand France’s Role in European Nuclear Defence

    February 25, 2026
    Facebook X (Twitter) Instagram
    Trending
    • Nvidia Breaks $215 Billion Revenue as AI Demand Drives Unprecedented Growth
    • Aston Martin to cut 20% of jobs after losses widen to £363.9m
    • Macron Plans to Expand France’s Role in European Nuclear Defence
    • Paramount Raises Warner Bros Bid, Escalating High-Stakes Clash With Netflix
    • US Consumer Confidence Rises in February
    • Gulf Allies Stand with Kuwait in Maritime Dispute with Iraq
    • Trump Rolls Out New Global Tariffs and Sparks Trade Tensions Worldwide
    • Government considers ban on unlicensed gambling sponsors in Premier League
    MirnewsMirnews
    • General
    • World
    • Finance
    • Money
    • Lifestyle
    Subscribe
    • News
    • Health
    • Media
    • Sports
    • Opinion
    • Real Estate
    • Education
    • Business & Economy
    • Entertainment
    • More
      • Travel & Tourism
      • Culture & Society
      • Environment & Sustainability
      • Technology & Innovation
      • Politics & Government
    MirnewsMirnews
    Home»Technology & Innovation»Long Conversations Weaken AI Safety
    Technology & Innovation

    Long Conversations Weaken AI Safety

    Rachel MaddowBy Rachel MaddowNovember 6, 2025No Comments2 Mins Read
    Facebook Twitter LinkedIn Telegram Pinterest Tumblr Reddit WhatsApp Email
    Share
    Facebook Twitter LinkedIn Pinterest Email

    AI systems lose their safety filters during longer chats, increasing the risk of harmful or inappropriate replies. A new report revealed that users can override safeguards in AI tools with just a few simple prompts.

    Cisco Tests Major Chatbots

    Cisco examined large language models from OpenAI, Mistral, Meta, Google, Alibaba, Deepseek, and Microsoft to determine how many prompts triggered unsafe information. Researchers ran 499 conversations using “multi-turn attacks,” where users asked several questions to slip past safety checks. Each chat contained between five and ten exchanges.

    The team compared responses from single and multiple prompts to see how easily chatbots shared dangerous or unethical data, such as private company information or misinformation. They extracted harmful content in 64 percent of multi-question sessions but only 13 percent of single-prompt ones.

    Success rates differed sharply, from 26 percent with Google’s Gemma to 93 percent with Mistral’s Large Instruct model. Cisco warned that these multi-turn methods could spread harmful content or grant hackers access to private data.

    Open Models Shift Safety Responsibility

    The study found that AI systems often forget their rules over longer conversations, letting attackers gradually adjust prompts and dodge safeguards. Mistral, Meta, Google, OpenAI, and Microsoft all use open-weight models, allowing the public to view their safety parameters. Cisco explained that these open systems usually contain lighter protections so users can modify them freely. This shifts safety responsibility to whoever customizes the model.

    Cisco noted that Google, OpenAI, Meta, and Microsoft have worked to limit malicious fine-tuning. However, AI developers still face criticism for weak guardrails that allow criminal misuse. In one case, U.S. firm Anthropic admitted that criminals used its Claude model for large-scale data theft and extortion, demanding ransoms exceeding $500,000 (€433,000).

    Share. Facebook Twitter Pinterest LinkedIn Tumblr Telegram Email
    Previous ArticleChelsea back Maresca’s rotation strategy despite Qarabag setback
    Next Article Tesla shareholders approve Elon Musk’s unprecedented $1 trillion compensation plan
    Rachel Maddow
    • Website
    • Facebook

    Rachel Maddow is a freelance journalist based in the USA, with over 20 years of experience covering Politics, World Affairs, Business, Health, Technology, Finance, Lifestyle, and Culture. She earned her degree in Political Science and Journalism from Stanford University. Throughout her career, she has contributed to outlets such as MSNBC, The New York Times, and The Washington Post. Known for her thorough reporting and compelling storytelling, Rachel delivers accurate and timely news that keeps readers informed on both national and global developments.

    Related Posts

    OpenAI Considered Alerting Police About Future Canadian School Shooter

    February 22, 2026

    Seattle Startup Funding Drives Tech Growth

    February 17, 2026

    Big Tech’s AI Spending Surge Threatens Europe’s Digital Independence

    February 16, 2026
    Add A Comment
    Leave A Reply Cancel Reply

    Latest News

    Government considers ban on unlicensed gambling sponsors in Premier League

    February 24, 2026

    EU Puts US Trade Deal on Hold Amid Legal Clash and New Tariffs

    February 23, 2026

    UK halts puberty blocker study as regulator calls for higher minimum age

    February 23, 2026

    The Trial That Could Change How Social Media Protects Young Users

    February 23, 2026

    Orange Juice Influences Gene Activity

    Health December 2, 2025

    Researchers show that daily orange juice alters thousands of immune-cell genes and affects vital body…

    Russian Cyberattacks Match Terror Threats in European Security Focus

    December 24, 2025

    Best Buy Boosts Full-Year Sales Outlook

    November 27, 2025

    Scientists Create First Accurate Blood Test for Chronic Fatigue Syndrome

    October 9, 2025

    Mir News brings you fresh stories, news, culture, and trends from the United States and beyond — your daily source for insight, inspiration, and authentic perspectives.

    We're social. Connect with us:

    Facebook Instagram
    Categories
    • Business & Economy
    • Culture & Society
    • Education
    • Entertainment
    • Environment & Sustainability
    • Health
    • Media
    • News
    • Opinion
    • Real Estate
    • Sports
    • Technology & Innovation
    • Travel & Tourism
    Latest News

    Devastating School Shooting Rocks Tumbler Ridge, B.C.

    February 11, 2026

    Maxwell Invokes Fifth Amendment as Lawmakers Press for Answers

    February 10, 2026

    ACC Halts European Battery Factory Plans Amid Slower EV Growth

    February 7, 2026
    All Rights Reserved © 2026 Mirnews.
    • Contact Us
    • Privacy Policy
    • Terms and conditions
    • Disclaimer
    • Imprint

    Type above and press Enter to search. Press Esc to cancel.