Since our February 2026 report on AI-related threat activity, Google Threat Intelligence Group (GTIG) has continued to track a maturing transition from nascent AI-enabled operations to the industrial-scale application of generative models within adversarial workflows. This report, based on insights derived from Mandiant incident response engagements, Gemini, and GTIG’s proactive research, highlights the dual nature of the current threat environment where AI serves as both a sophisticated engine for adversary operations and a high-value target for attacks. We explore the following developments: Vulnerability Discovery and Exploit Generation; AI-Augmented Development for Defense Evasion; Autonomous Malware Operations; AI-Augmented Research and IO: Obfuscated LLM Access; Supply Chain Attacks.
Read more…
Source: Google Threat Intelligence Group
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- Consumers Urged to Junk Insecure IoT Devices
June 18, 2019
More than 2 million connected security cameras, baby monitors and other IoT devices have serious vulnerabilities that have been publicly disclosed for more than two months – yet they are still without a patch or even any vendor response. Security researcher Paul Marrapese, who disclosed the flaws in April and has yet to hear back from any impacted ...
- Mobile Cyberespionage Campaign ‘Bouncing Golf’ Affects Middle East
June 18, 2019
We uncovered a cyberespionage campaign targeting Middle Eastern countries. We named this campaign “Bouncing Golf” based on the malware’s code in the package named “golf.” The malware involved, which Trend Micro detects as AndroidOS_GolfSpy.HRX, is notable for its wide range of cyberespionage capabilities. Malicious codes are embedded in apps that the operators repackaged from legitimate ...
- Plurox: Modular backdoor
June 18, 2019
In February this year, a curious backdoor passed across our virtual desk. The analysis showed the malware to have a few quite unpleasant features. It can spread itself over a local network via an exploit, provide access to the attacked network, and install miners and other malicious software on victim computers. What’s more, the backdoor ...
- Houdini malware targets victims with keylogger, online bank account theft tools
June 17, 2019
A new variant of the Houdini malware has been detected in campaigns against financial institutions and their customers. Last week, cybersecurity researchers from Cofense said in a blog post that the new strain of Houdini — also known as HWorm — was released by its author on June 2, 2019. Dubbed WSH Remote Access Tool (RAT), it took the ...
- New Echobot malware is a smorgasbord of vulnerabilities
June 17, 2019
If there’s one thing that seems to have no end in sight is malware authors putting their own spin on the old Mirai malware and creating new botnets to haunt the IoT and enterprise landscapes. Not a month goes by without a new major botnet appearing out of nowhere and launching massive attacks against people’s smart ...
- AESDDoS Botnet Malware Infiltrates Containers via Exposed Docker APIs
June 14, 2019
Misconfiguration is not novel. However, cybercriminals still find that it is an effective way to get their hands on organizations’ computing resources to use for malicious purposes and it remains a top security concern. In this blog post, we will detail an attack type where an API misconfiguration in the open-source version of the popular DevOps tool ...