All posts by FreeNews

United States: Governor Abbott Announces $113 Million In Grants For Electric Reliability In The Panhandle And South Plains

Governor Greg Abbott today announced $113 million in Texas Energy Fund (TxEF) grants to strengthen reliability for Southwestern Public Service Company’s (SPS) customers in the Texas Panhandle and South Plains. SPS is a subsidiary of Xcel Energy. The grants will help fund projects to repair or replace power poles, increase protections in higher-risk areas of […]

United States: URIF Holds Fourth Board Meeting At Ukraine Recovery Conference

The U.S.–Ukraine Reconstruction Investment Fund (URIF) held its fourth board meeting on Wednesday, June 24, on the sidelines of the Ukraine Recovery Conference in Gdańsk, Poland. At the meeting, URIF’s Board of Directors discussed the landmark cooperation agreement between the U.S. International Development Finance Corporation (DFC) and the Multilateral Investment Guarantee Agency (MIGA) – the […]

United States: U.S. Department Of Education Partners With The U.S. Department Of Justice To Protect Parental Rights

Today, the U.S. Department of Education (the Department) announced it will work alongside the U.S. Department of Justice (DOJ) to protect parental rights in Kansas. In April 2026, the Department’s Student Privacy Policy Office (SPPO) found that Kansas City, Kansas Public School District (the District) has policies that violate the Family Educational Rights and Privacy Act […]

United States: United States-Tajikistan Annual Bilateral Consultations

Today, the State Department held a successful round of the United States-Tajikistan Annual Bilateral Consultations, the first under the Trump Administration and a meaningful restart of high-level dialogue following a four-year pause. U.S. Assistant Secretary of State for South and Central Asian Affairs Paul Kapur recognized Tajikistan’s key role in border security efforts aimed at […]

New attack provides one more reason why AI browsers are a bad idea

Makers of AI browsers make lofty promises. With a single prompt, users can ask one to find a restaurant in a particular part of town, reserve a table, invite a colleague to lunch, and email a confirmation. These makers are much more reticent about the risks of blurring the once fine line between browsing sites and asking a large language model a question or instructing it to take potentially sensitive actions.

LLM developers’ answer so far has been to build guardrails that make some requests off-limits. Developing software exploits, stealing credentials, or teaching how to build a pipe bomb are examples. The problem with this approach is that the guardrails are reactive and treat the symptoms rather than solve the root cause. It’s tantamount to the manufacturer of an unsafe vehicle advocating for new road designs rather than fixing the flaws that make it prone to accidents.

Lulling LLMs into an alternate reality

New research puts this predicament on sharp display. It demonstrates how a website can lull AI browsers into a false reality where the rules governing its behavior no longer apply. After that, an attacker has free rein to invoke all kinds of destructive actions, such as extracting code from a private repository or extracting credentials from the built-in password manager.

Read full article

Comments