Close Menu

    Subscribe to Updates

    What's Hot

    Bitcoin liquidation map flags $65,000 as key support, $68,000 as squeeze zone

    April 3, 2026

    CFTC and DOJ sue three states over prediction market oversight

    April 3, 2026

    DeepMind flags six web based attacks that can hijack AI agents

    April 3, 2026
    Facebook X (Twitter) Instagram
    laicryptolaicrypto
    Demo
    • Ethereum
    • Crypto
    • Altcoins
    • Blockchain
    • Bitcoin
    • Lithosphere News Releases
    laicryptolaicrypto
    Home DeepMind flags six web based attacks that can hijack AI agents
    Crypto

    DeepMind flags six web based attacks that can hijack AI agents

    John SmithBy John SmithApril 3, 2026No Comments3 Mins Read
    Share
    Facebook Twitter LinkedIn Pinterest Email



    Researchers at Google DeepMind have warned that the open internet can be used to manipulate autonomous AI agents and hijack their actions.

    Summary

    • DeepMind researchers have identified six attack methods that can be used to manipulate autonomous AI agents as they browse and act online.
    • The study warned that hidden instructions, persuasive language, and poisoned data sources can influence agent decisions or override safeguards.

    The study titled “AI Agent Traps” comes as companies deploy AI agents for real-world tasks and attackers begin using AI for cyber operations.

    Instead of focusing on how models are built, the research looks at the environments agents operate in. It identifies six types of traps that take advantage of how AI systems read and act on information from the web.

    The six attack categories outlined in the paper include content injection traps, semantic manipulation traps, cognitive state traps, behavioural control traps, systemic traps, and human in the loop traps.

    Content injection stands out as one of the most direct risks. Hidden instructions can be placed inside HTML comments, metadata, or cloaked page elements, allowing agents to read commands that remain invisible to human users. Tests showed these techniques can take control of agent behaviour with high success rates.

    Semantic manipulation works differently, relying on language and framing rather than hidden code. Pages loaded with authoritative phrasing or disguised as research scenarios can influence how agents interpret tasks, sometimes slipping harmful instructions past built-in safeguards.

    Another layer targets memory systems. By planting fabricated information into sources that agents rely on for retrieval, attackers can influence outputs over time, with the agent treating false data as verified knowledge.

    Behavioural control attacks take a more direct route by targeting what an agent actually does. In these cases, jailbreak instructions can be embedded into normal web content and read by the system during routine browsing. Separate tests showed that agents with broad access permissions could be pushed into locating and transmitting sensitive data, including passwords and local files, to external destinations.

    System-level risks extend beyond individual agents, with the paper warning that coordinated manipulation across many automated systems could trigger cascading effects, similar to past market flash crashes driven by algorithmic trading loops.

    Human reviewers are also part of the attack surface, as carefully crafted outputs can appear credible enough to gain approval, allowing harmful actions to pass through oversight without raising suspicion.

    How to defend against these risks?

    To counter these risks, researchers suggest a mix of adversarial training, input filtering, behavioural monitoring, and reputation systems for web content. They also point to the need for clearer legal frameworks around liability when AI agents execute harmful actions.

    The paper stops short of offering a complete fix and argues that the industry still lacks a shared understanding of the problem, leaving current defenses scattered and often focused on the wrong areas.



    Source link

    Share. Facebook Twitter Pinterest LinkedIn WhatsApp Reddit Tumblr Email
    John Smith

    Related Posts

    Bitcoin liquidation map flags $65,000 as key support, $68,000 as squeeze zone

    April 3, 2026

    CFTC and DOJ sue three states over prediction market oversight

    April 3, 2026

    Asia markets rise led by Nikkei and Kospi, how are crypto equities faring?

    April 3, 2026
    Leave A Reply Cancel Reply

    Demo
    Don't Miss
    Crypto

    Bitcoin liquidation map flags $65,000 as key support, $68,000 as squeeze zone

    By John SmithApril 3, 20260

    Coinglass’ Bitcoin liquidation map shows a $1.143b long wall below $65k and a $754m short…

    CFTC and DOJ sue three states over prediction market oversight

    April 3, 2026

    DeepMind flags six web based attacks that can hijack AI agents

    April 3, 2026

    Lithosphere Advances AI-Native Blockchain Infrastructure with Makalu Testnet and Integrated Protocol Stack

    April 3, 2026

    LAI Crypto is a user-friendly platform that empowers individuals to navigate the world of cryptocurrency trading and investment with ease and confidence.

    Our Posts
    • Altcoins (33)
    • Bitcoin (10)
    • Blockchain (9)
    • Crypto (724)
    • Ethereum (487)
    • Lithosphere News Releases (19)

    Subscribe to Updates

    • Twitter
    • Instagram
    • YouTube
    • LinkedIn

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