China's AI Gap: A Reality Check in the Face of Growing Tensions A recent study by the UK Artificial Intelligence Security Institute (AISI) and the US Centre for AI Standards and Innovation (CAISI) has shed light on a concerning trend in artificial intelligence.
The report evaluated China's Kimi K3 model against top American rivals, revealing that Beijing's most powerful large language model falls significantly short in its ability to launch cyberattacks.
The study found that Kimi K3 scored 32. 2% on the ExploitBench benchmark, a far cry from the top US models that averaged an impressive 76. 2%. This gap is not marginal; it's significant.