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.
These attacks could manipulate the browser or steal private information through hidden instructions.
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 note: Beware of AI-powered browsers' security risks.
More News
- Guardians Muddled by Gambling Scandals
- BMA slams ‘indefensible’ pay offer
- 'Argentina gradually joins the great authoritarian wave sweeping the globe'
- Do Trump Drug Boat Killings Expose US Servicemembers to Criminal Liability? [FEATURE] | StoptheDrugWar.org
- The power of organising and how to win the outsourcing battle | Article | News | UNISON National
- BMA says Government’s suggested pay rise for doctors in 2026 is indefensible - BMA media centre - BMA
- 'BHV has muddled its own narrative by associating with Shein'
- United States: Organising to resist mass immigrant raids