Whispering poetry at AI can make it break its own rules

Most of the big AI makers don't like people using their models for unsavory activity. Ask one of the mainstream AI models how to make a bomb or create nerve gas and you'll get the standard "I don't help people do harmful things" response. That has spawned a cat-and-mouse game of people who try to...

Using LLMs to Create Bioweapons

Im not sure there are good ways to build guardrails to prevent this sort of thing: There is growing concern regarding the potential misuse of molecular machine learning models for harmful purposes. Specifically, the dual-use application of models for predicting cytotoxicity18 to create new poison...

EU sanctions hackers from China, Russia, North Korea who're wanted by the FBI

The Council of the European Union has imposed its first-ever sanctions against persons or entities involved in various cyber-attacks targeting European citizens, and its member states. The directive has been issued against six individuals and three entities responsible for or involved in various...

Laser Gun to Detect Bombs and Chemical Weapons from 100 Feet Away

In Brief US-based Intelligence Advanced Research Projects Activity IARPA agency is sponsoring a program to build portable laser sensors that could detect explosives, narcotics and other dangerous chemical weapons from 100 feet away. Dubbed SILMARILS Standoff Illuminator for Measuring Absorbance a...

Email Spam Claims US Attacks Syria and Leads to Malware

A new phishing campaign is disseminating malicious links with emails purporting to come from CNN saying that the United States has initiated military strikes against the embattled regime of Syrian President Bashar al Assad. One such email, obtained by Kaspersky Lab and posted on Securelist, comes...