Close Menu
GeekBlog

    Subscribe to Updates

    Get the latest creative news from FooBar about art, design and business.

    What's Hot

    Magic: The Gathering PAX Panel Previews Seriously Sinister Supervillains

    August 30, 2025

    Meta will sell you refurbished Ray-Ban smart glasses for $76 off – how to find them

    August 30, 2025

    Garmin Fenix 8 Pro rumors swirl, and new leaks point to 4 new subscription tiers – mere months after the Connect+ debacle

    August 30, 2025
    Facebook X (Twitter) Instagram Threads
    GeekBlog
    • Home
    • Mobile
    • Reviews
    • Tech News
    • Deals & Offers
    • Gadgets
      • How-To Guides
    • Laptops & PCs
      • AI & Software
    • Blog
    Facebook X (Twitter) Instagram
    GeekBlog
    Home»Mobile»This AI didn’t just simulate an attack – it planned and executed a real breach like a human hacker
    Mobile

    This AI didn’t just simulate an attack – it planned and executed a real breach like a human hacker

    Michael ComaousBy Michael ComaousAugust 2, 2025No Comments3 Mins Read0 Views
    Share Facebook Twitter Pinterest LinkedIn Tumblr Email Copy Link
    A laptop with digitally inserted hack warnings around it
    Share
    Facebook Twitter LinkedIn Pinterest Email Copy Link

    • Researchers recreated the Equifax hack and watched AI do everything without direct control
    • The AI model successfully carried out a major breach with zero human input
    • Shell commands weren’t needed, the AI acted as the planner and delegated everything else

    Large language models (LLMs) have long been considered useful tools in areas like data analysis, content generation, and code assistance.

    However, a new study from Carnegie Mellon University, conducted in collaboration with Anthropic, has raised difficult questions about their role in cybersecurity.

    The study showed that under the right conditions, LLMs can plan and carry out complex cyberattacks without human guidance, suggesting a shift from mere assistance to full autonomy in digital intrusion.


    You may like

    From puzzles to enterprise environments

    Earlier experiments with AI in cybersecurity were mostly limited to “capture-the-flag” scenarios, simplified challenges used for training.

    The Carnegie Mellon team, led by PhD candidate Brian Singer, went further by giving LLMs structured guidance and integrating them into a hierarchy of agents.

    With these settings, they were able to test the models in more realistic network setups.

    In one case, they recreated the same conditions that led to the 2017 Equifax breach, including the vulnerabilities and layout documented in official reports.

    Sign up to the TechRadar Pro newsletter to get all the top news, opinion, features and guidance your business needs to succeed!

    The AI not only planned the attack but also deployed malware and extracted data, all without direct human commands.

    What makes this research striking is how little raw coding the LLM had to perform. Traditional approaches often fail because models struggle to execute shell commands or parse detailed logs.

    Instead, this system relied on a higher-level structure where the LLM acted as a planner while delegating lower-level actions to sub-agents.

    This abstraction gave the AI enough context to “understand” and adapt to its environment.

    Although these results were achieved in a controlled lab setting, they raise questions about how far this autonomy could go.

    The risks here are not just hypothetical. If LLMs can carry out network breaches on their own, then malicious actors could potentially use them to scale attacks far beyond what’s feasible with human teams.

    Even tools such as endpoint protection and the best antivirus software may be tested by such adaptive and responsive agents.

    Nevertheless, there are potential benefits to this capability. An LLM capable of mimicking realistic attacks might be used to improve system testing and expose flaws that would otherwise go unnoticed.

    “It only works under specific conditions, and we do not have something that could just autonomously attack the internet… But it’s a critical first step,” said Singer in explaining that this work remains a prototype.

    Still, the ability of an AI to replicate a major breach with minimal input should not be dismissed.

    Follow-up research is now exploring how these same techniques can be applied in defense, potentially even enabling AI agents to detect or block attacks in real-time.

    You may also like

    attack breach didnt executed hacker human planned real simulate
    Follow on Google News Follow on Flipboard
    Share. Facebook Twitter Pinterest LinkedIn Tumblr Telegram Email Copy Link
    Previous ArticleBest Stores for Buying MP3 and Digital Music You Can Keep Forever
    Next Article 11 Best Coolers WIRED Tested for Every Budget, Any Situation (2025)
    Michael Comaous
    • Website

    Related Posts

    5 Mins Read

    A little of that human touch: rethinking CX in the Age of AI

    3 Mins Read

    A disgruntled worker built his own kill-switch malware to take down his former employer – and it didn’t pay off

    3 Mins Read

    College student’s “time travel” AI experiment accidentally outputs real 1834 history

    3 Mins Read

    Coinbase CEO explains why he fired engineers who didn’t try AI immediately

    3 Mins Read

    Arch Linux remains under attack as DDoS enters week 2 – here’s a workaround

    2 Mins Read

    Lenovo’s Lena AI chatbot could be turned into a secret hacker with just one question

    Top Posts

    8BitDo Pro 3 review: better specs, more customization, minor faults

    August 8, 202512 Views

    WIRED Roundup: ChatGPT Goes Full Demon Mode

    August 2, 202512 Views

    Framework Desktop Review: A Delightful Surprise

    August 7, 202511 Views
    Stay In Touch
    • Facebook
    • YouTube
    • TikTok
    • WhatsApp
    • Twitter
    • Instagram
    Latest Reviews

    Subscribe to Updates

    Get the latest tech news from FooBar about tech, design and biz.

    Most Popular

    8BitDo Pro 3 review: better specs, more customization, minor faults

    August 8, 202512 Views

    WIRED Roundup: ChatGPT Goes Full Demon Mode

    August 2, 202512 Views

    Framework Desktop Review: A Delightful Surprise

    August 7, 202511 Views
    Our Picks

    Magic: The Gathering PAX Panel Previews Seriously Sinister Supervillains

    August 30, 2025

    Meta will sell you refurbished Ray-Ban smart glasses for $76 off – how to find them

    August 30, 2025

    Garmin Fenix 8 Pro rumors swirl, and new leaks point to 4 new subscription tiers – mere months after the Connect+ debacle

    August 30, 2025

    Subscribe to Updates

    Get the latest creative news from FooBar about art, design and business.

    Facebook X (Twitter) Instagram Pinterest Threads
    • About Us
    • Contact us
    • Privacy Policy
    • Disclaimer
    • Terms and Conditions
    © 2025 geekblog. Designed by Pro.

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