risk assessment
-
AI Safety Benchmark: Code Model Safety Testing Results Released
CAICT’s AI Institute launched security benchmark testing for code-generating LLMs, assessing risks and capabilities using a dataset of 15,000+ test cases across nine languages and various attack methods. The initial assessment of 15 Chinese models (3B-671B parameters) revealed varied security levels, with most exhibiting medium risk. Models showed weaknesses in scenarios involving malicious intent, highlighting vulnerabilities to cyberattacks. CAICT plans to expand testing to international models and develop mitigation tools, aiming to promote a secure LLM ecosystem.
-
Meta Plans to Fully Automate Ad Creative with AI: Images, Videos, and Text
Meta plans to fully automate AI-powered ad creation for brands by the end of next year, leveraging AI to generate visuals, videos, and text based on product images and budget goals. This initiative aims to further strengthen Meta’s advertising business, which accounts for over 97% of its revenue. Simultaneously, Meta is shifting towards AI for risk assessment, targeting AI to handle up to 90% of related tasks.