Researchers warn AI agents could become botnets through hallucinations

The post Ethereum Tests AI Agents on Protocol Code, Finds Critical Bugs appeared on BitcoinEthereumNews.com. Jessie A Ellis Jul 09, 2026 16:11 Ethereum Foundation runs AI agents on protocol code, uncovering critical bugs like CVE-2026-34219 in libp2p. Here’s why this matters for Ethereum’s future. The Ethereum Foundation’s Protocol Security team has revealed a cutting-edge approach to enhancing the network’s integrity: deploying AI agents to analyze Ethereum’s protocol code. In a new blog post published on July 9, 2026, the team detailed how these agents identified critical vulnerabilities, including a remotely-triggerable bug in libp2p’s gossipsub layer, now patched and disclosed as CVE-2026-34219. Ethereum (ETH), trading at $1,624.95 as of July 9, remains the backbone of decentralized finance (DeFi) and smart contract platforms. Protecting its base layer is essential for maintaining trust in the network, which secures billions of dollars in economic activity daily.

A security hole within AI dev tools has allowed attackers to escape sandboxes by misleading the humans in the loop who were supposed to knowingly approve the tool’s actions, according to cybersecurity research firm Wiz. “We discovered GhostApproval, a systematic vulnerability pattern affecting six of the top AI coding assistants: Amazon Q Developer, Anthropic Claude Code, Augment, Cursor, Google Antigravity, and Windsurf [now known as Devin Desktop],” the Wiz report said. “In each case, a malicious repository can trick the agent into accessing arbitrary files outside the workspace sandbox, potentially achieving remote code execution on the developer’s machine.” The first report of the hole came earlier this month from Cato Networks, but was limited to one platform, Cursor, whereas Wiz found that its impact was far wider.  The underlying security problem, symbolic links (symlinks), is well known and has been leveraged for decades. But GhostApproval, Wiz noted, goes well beyond their his

OpenAI's latest family of models promise improvements across a range of areas, including cybersecurity.

Lyzr, a startup that builds AI agents for enterprises, used its own AI agent to raise a $100 million round — proof, evidently, that the product actually works.

Widespread API key sharing among AI agents heightens security risks, potentially leading to severe breaches and irreversible damages, especially in crypto. The post 69% of enterprises share API keys among AI agents, creating massive security blind spots appeared first on Crypto Briefing.

The post AI Agents Could Be Turned Into Botnets Through Hallucinations, Researchers Warn appeared on BitcoinEthereumNews.com. In brief Researchers introduced “Adversarial HalluSquatting,” an attack that exploits AI-generated hallucinations. The technique tricks AI agents into trusting fake repositories or tools that contain malicious instructions. Tests against popular AI coding assistants showed the method could lead to remote code execution in controlled experiments. AI hallucinations may be more than incorrect answers—they could become a way for hackers to compromise computers, according to new research from Tel Aviv University, Technion, and Intuit. In the paper, “Beware of Agentic Botnets: Scalable Untargeted Promptware Attacks via Universal and Transferable Adversarial HalluSquatting,” researchers demonstrated a technique that exploits AI models when they generate fake links to software repositories and other online resources.  “The growing adoption of agentic LLM applications h

Researchers warn AI agents could be tricked into downloading malicious code by exploiting the same hallucinations that cause chatbots to make mistakes.

Ethereum Foundations Protocol Security team said AI agents can help find real bugs in protocol code, but warned that triage, reproducibility, and human review remain the core of security work. The post Ethereum Foundation says AI agents can find real bugs but triage is the real work appeared first on Crypto Briefing.