Be Cautious with Agentic Web Browsers
New AI-powered browsers from OpenAI, Perplexity, and others promise to automate Web tasks but are vulnerable to prompt injection attacks.
While they are intriguing for occasional experiments, it is recommended to stick with your current browser for everyday usage due to potential security risks.
Researchers have uncovered vulnerabilities in Atlas allowing attackers to take advantage of ChatGPT’s “memory” to inject malicious code, grant themselves access privileges, or deploy malware.
- Flaws discovered in Comet could allow attackers to hijack the browser’s AI with hidden instructions.
- Perplexity and OpenAI’s chief information security officer, Dane Stuckey, acknowledged prompt injections as a big threat.
Author's summary: Beware of AI-powered browser vulnerabilities.
More News
- Kangana Ranaut’s Farmer Post Row Becomes A Study In Free Speech And Public Sensitivity
- A tale of two theaters: David Miller’s path to success in the local arts - The Parthenon
- India chase on point, level series against Australia
- Cindy Crawford and Kaia Gerber shine as mother-daughter duo at 2025 LACMA gala | The Express Tribune