Digit.in

Google says era of autonomous attacks is here, warns AI helping hackers to exploit software bugs

Highlights

Hackers are using AI tools to automate cyberattacks and exploit software weaknesses.

Criminals are stealing AI subscriptions and cloud resources to make money.

AI is also being used to create fake identities and spread misleading content online.

Google says era of autonomous attacks is here, warns AI helping hackers to exploit software bugs

Google recently pointed out that the threat actors are getting more and more advanced as they are now using artificial intelligence to automate hacking, exploit software weaknesses and run large-scale online influence campaigns. A new report by Google Threat Intelligence Group (GTIG) and Mandiant highlights how hackers are moving beyond basic AI prompts to autonomous systems that can carry out multiple stages of an attack with limited human involvement. They also talk about how hackers are stealing access to advanced AI tools, hijacking cloud computing resources, and creating fake online identities. Here is everything Google said in its report.

Hackers use AI agents to automate cyberattacks

The latest report by Google Threat Intelligence Group (GTIG) and Mandiant identifies growing use of open-source AI-powered offensive frameworks such as ARTEX, CAIRN, CyberStrikeAI and PentAGI. The report stated that these tools can help attackers identify weaknesses, generate code and adjust their actions based on real-time results.

Google also observed that CAIRN and PentAGI were being used in real-world operations, while other frameworks remained in the testing phase. It further noted that in one campaign, attackers used a customised version of PentAGI to scan organisations throughout Asia.

AI makes vulnerability exploitation faster

Google says era of autonomous attacks is here, warns AI helping hackers to exploit software bugs

Google said that the threat actors are increasingly using AI to study software updates, identify security gaps and develop exploits. Moreover, the report also describes an automated system called exploit_agent that compares software versions before and after security patches, identifies possible vulnerabilities and attempts to develop and test exploits.

As a result, attackers can attempt to exploit vulnerabilities much sooner after they become public, although AI-generated exploits may not always work, creating additional work for security teams.

LLMJacking turns stolen AI access into a business

The report from Google Threat Intelligence Group (GTIG) and Mandiant also noted that the attackers are increasingly using LLMJacking, which is a method to steal cloud credentials and misuse victims' computing resources and AI subscriptions.

It further states that in some incidents this resulted in fraudulent charges exceeding $100,000 in a single day, with the most severe cases reaching $1 million. Not only this but the criminals are also selling unauthorised access through underground services known as LLM Mixers, allowing buyers to use stolen AI accounts and computing resources.

AI fuels fake news and online influence campaigns

If this was not already enough, the report also highlighted that the state-backed groups, commercial influence firms and financially motivated networks are also using AI to create fake identities, generate political content and distribute it across social media. Google also found attempts to build platforms which are capable of managing more than 1,00,000 accounts using automated personas and AI-generated content. However, Google noted that better content quality does not necessarily mean greater impact.

Google's next move

To protect consumers, Google said it uses threat intelligence, security monitoring, and AI-based tools to find and stop harmful activities. Its defence strategy also includes improving model security, disabling accounts involved in misuse, and using automated tools to find security weaknesses.

Bhaskar Sharma

Bhaskar Sharma

Bhaskar Sharma Bhaskar is a Senior Copy Editor at Digit India who keeps a close watch on everything shaping the world of technology from smartphones and home appliances to AI, government tech initiatives, digital safety, and the latest industry developments. Whether it's breaking news, in-depth features, hands-on reviews, practical how-to guides, or exclusive scoops, he translates complex tech into stories that are easy to understand and worth reading. His work has been featured in iGeeksBlog, GuidingTech, and other leading publications. Before joining Digit India, he served as an assistant editor at TechBloat. A B.Tech graduate and full-time tech journalist, he is driven by just one goal, which is to help readers stay informed, stay secure, and stay ahead in an ever-changing digital world.

View full profile