{"id":1383063,"date":"2025-11-06T12:38:03","date_gmt":"2025-11-06T12:38:03","guid":{"rendered":"http:\/\/2AYQBg5XVsFGMFjLv2okj5"},"modified":"2025-11-06T12:38:03","modified_gmt":"2025-11-06T12:38:03","slug":"great-now-even-malware-is-using-llms-to-rewrite-its-code-says-google-as-it-documents-new-phase-of-ai-abuse","status":"publish","type":"post","link":"https:\/\/arcader.org\/news\/great-now-even-malware-is-using-llms-to-rewrite-its-code-says-google-as-it-documents-new-phase-of-ai-abuse\/","title":{"rendered":"Great, now even malware is using LLMs to rewrite its code, says Google, as it documents new phase of ‘AI abuse’"},"content":{"rendered":"
\n

AI has been up to some naughty things in the past few years. From ignoring copyright<\/a> to producing pretty awful approximations of games<\/a>, to being exploited to lock down your machine<\/a>. It seems like, as AI gets smarter, it only gets easier to use in malicious ways. One piece of malware has even been spotted using AI to rewrite itself to avoid detection. Spooky stuff.<\/p>\n

Google has recently warned<\/a> “adversaries are no longer leveraging artificial intelligence (AI) just for productivity gains, they are deploying novel AI-enabled malware in active operations”.<\/p>\n

The report (via Bleeping Computer<\/a>) says that, for the first time, Google “discovered a code family that employed AI capabilities mid-execution to dynamically alter the malware\u2019s behaviour.”<\/p>\n

Google pays attention to one particular malware named Promptflux, which is a dropper that prompts Gemini to rewrite its own source code, “saving the new, obfuscated version to the Startup folder to establish persistence.” Like an actual virus, this suggests it has the ability to ‘evolve’ in some way, getting smarter and harder to get rid of. It is only experimental right now and doesn’t appear to be “used in the wild”, luckily.<\/p>\n

If you’re worried about the potential of this tech (same), Google says DeepMind has used insights gained from the malware to strengthen security, both with “Google\u2019s classifiers and the model itself.” Reportedly, this should mean that the model refuses to help malware with these kinds of attacks going forward. AI fighting AI. So it begins.<\/p>\n

\n
\n
\n

\"Fallout<\/p>\n<\/div>\n<\/div>

(Image credit: Bethesda)<\/span><\/figcaption><\/figure>\n

Naturally, AI being so broad and having so many applications does often mean it’s exploitable, and it wouldn’t be the first time someone has got AI to act against its reported purpose<\/a>.<\/p>\n

In its latest report, Google has identified four other cases of AI malware that have popped up in 2025. They are:<\/p>\n