logo
About Contact

Chinese AI Models Demonstrate Ability to Execute Sophisticated Cyberattacks in Tests

AI News

Chinese AI Models Demonstrate Ability to Execute Sophisticated Cyberattacks in Tests

Chinese AI model carries out complex cyberattacks in controlled tests, researchers say

Economic Times

Chinese AI model carries out complex cyberattacks in controlled tests, researchers say

Chinese firm Z.AI's flagship AI model carried out complex cyberattacks in controlled simulations and repeatedly ignored instructions in separate tests, according to a report by AI security...

Chinese AI model carries out complex cyberattacks in controlled tests, researchers say

Reuters

Chinese AI model carries out complex cyberattacks in controlled tests, researchers say

Chinese firm Z.AI's flagship AI model carried out ​complex cyberattacks in controlled simulations and repeatedly ignored instructions in separate tests, according to a report by AI security...

Chinese AI models can execute complex cyberattacks, research reveals

NewsBytes

Chinese AI models can execute complex cyberattacks, research reveals

Chinese firm Z.AI's flagship AI model executed complex cyberattacks during controlled simulations, as reported by AI security research firm Irregular.

GLM-5.3's Cyberattack Capabilities in AI Testing Raise Concerns Over Security

GLM-5.3 Completes Cyberattacks in AI Tests

Technology Org

GLM-5.3 Completes Cyberattacks in AI Tests

Irregular says Z.ai's open-weight GLM-5.3 finished 3 of 13 multi-stage cyberattack scenarios and ignored instructions in its tests.

NewsBytes

Irregular report finds GLM-5.3 hacked systems, trails OpenAI's GPT-6 Astra

Chinese firm Z.AI's GLM-5.3 model successfully completed complex cyberattack scenarios in controlled tests, raising concerns about the potential misuse of open-weight AI systems.

Chinese AI model completes complex cyberattacks in tests, raises concerns

Business Standard

Chinese AI model completes complex cyberattacks in tests, raises concerns

Z.AI's GLM-5.3 completed three of 13 multi-stage cyberattack scenarios in controlled simulations, according to Irregular, though it trailed leading US models in separate tests

Coverage dates

First reported:

First saved on ClustZ:

Coverage updated:

We use cookies to improve your experience and analyze site traffic. By clicking "Accept", you agree to our use of cookies. See our Privacy Policy.