Gate News message: Researchers at Google DeepMind warn that an open internet environment could be leveraged to hijack autonomous AI agents and manipulate their behavior. The report, titled “AI Agent Traps,” states that when companies deploy AI agents to carry out real tasks, attackers may also launch targeted attacks over the network. The study identifies six major risks, including content injection traps, semantic manipulation traps, cognitive state traps, behavior control traps, system traps, and human-agent interaction traps.
The content injection trap is the most direct: attackers can place instructions in HTML comments, metadata, or hidden page elements, which the agent can read and then execute. Semantic manipulation traps work by loading authoritative phrasing or by disguising themselves as webpages in a research environment, quietly affecting the agent’s understanding of the task—sometimes even bypassing safety mechanisms. Cognitive state traps work by implanting false data into the agent’s information sources, causing it to mistakenly believe for the long term that this information has been verified. Behavior control traps target the agent’s actual operations, potentially luring it to access sensitive data and transmit it to an external target.
System traps involve coordinated manipulation across multiple AI systems, which could trigger cascading effects—similar to how algorithmic trading can cause sudden market crashes. Human-agent interaction traps exploit human review steps by creating seemingly credible review content, allowing harmful behavior to slip past oversight.
To address these risks, DeepMind recommends combining adversarial training, input filtering, behavior monitoring, and network content reputation systems, while also establishing a clearer legal responsibility framework. However, the study notes that the industry still lacks unified defense standards, and that existing measures are often fragmented and focused differently. The study calls on developers and businesses to pay attention to operational environment security for AI agents to prevent potential risks of network manipulation and abuse.
Disclaimer: The information on this page may come from third parties and does not represent the views or opinions of Gate. The content displayed on this page is for reference only and does not constitute any financial, investment, or legal advice. Gate does not guarantee the accuracy or completeness of the information and shall not be liable for any losses arising from the use of this information. Virtual asset investments carry high risks and are subject to significant price volatility. You may lose all of your invested principal. Please fully understand the relevant risks and make prudent decisions based on your own financial situation and risk tolerance. For details, please refer to
Disclaimer.
Related Articles
Sanctioned Exchange Grinex Hit by $13.7M Hack; Blames Foreign Intelligence Services
Grinex, a sanctioned crypto-ruble exchange, has halted operations due to a cyberattack that stole over $13.74 million in USDT. The attack is believed to involve state-level actors aiming to destabilize Russia's financial system. Grinex is cooperating with law enforcement but has no timeline for resuming services.
Coinpedia7h ago
Figure Faces Short Seller Accusations Over Blockchain Integration Claims; FIGR Stock Down 53% From January Peak
Figure Technology Solutions faced allegations from Morpheus Research of overstating its blockchain technology use, resulting in a significant drop in share prices. Figure defended its operations, highlighting its digital asset features and strong performance metrics.
GateNews14h ago
Houston Crypto Fraudster Sentenced to 23 Years for $20M Meta-1 Coin Scam
Robert Dunlap, a Houston entrepreneur, was sentenced to 23 years in prison for a $20 million cryptocurrency fraud involving fake assets and deceptive practices, impacting over 1,000 victims. His case reflects a broader rise in crypto-related cybercrimes.
GateNews19h ago
SlowMist Warns of Active Phishing Attack Using Fake 'Harmony Voice' Software
SlowMist's security team has warned of a social engineering campaign targeting cryptocurrency users. Fraudsters are posing as project partners to trick users into downloading a malicious application disguised as a translation tool. Users are advised to verify software authenticity.
GateNews20h ago
Zonda Exchange CEO Blames Missing Founder for $336M in Lost Bitcoin
Zonda CEO Przemysław Kral has attributed the exchange's loss of access to 4,500 BTC, valued at $336 million, to missing founder Sylwester Suszek's failure to transfer private keys. Amid allegations of bankruptcy and intensified withdrawal requests, Kral insists Zonda remains solvent and will pursue legal action while searching for Suszek, who disappeared in 2022.
GateNews20h ago
Grinex Exchange Halts All Trading After $15M Cyberattack on Wallet Systems
Grinex, a Kyrgyz crypto exchange, suspended trading after a cyberattack resulting in losses of around $15 million. The advanced nature of the attack points to organized or state-level involvement. Grinex has reported the incident to authorities and is assessing the damage.
GateNews20h ago
