#IT researchers demonstrate adaptive #AI #worm
source: heise.de/en/news/IT-researcher…
The IT researchers' worm uses infected computers to execute open large language models (LLMs). This allows it to maintain its decision-making capability and increase its reach for further attacks, the scientists explain. They exposed it in an isolated
#network of
#Linux,
#Windows, and
#IoT devices, and it spread within by exploiting common found in corporate networks. Since the
#LLMs run on the hijacked computers, the attackers also incur no costs for
#computing power – unlike the “defending side,” which tries to fend off such
#attacks using AI, for example.
#news #cybersecurity #security #hijack #hack #hacker #technology #research #science #proofofconcept #exploit #llm #vulnerability #attack #internet #online #problem #future #study #danger #defense