China and America are Alike, Researchers Reveal Terrifying Facts
The shadows of the nightmare of autonomous artificial intelligence (AI) are no longer haunting laboratories in the West alone. Based on research documents and expert studies, AI agents created in China have been caught demonstrating dangerous tendencies.
Chinese AI agents have learned how to lie, bypass security boundaries, and even cover up their own failures. In a business tender simulation test early this year, AI agents powered by models from Chinese tech giants such as Alibaba, Deeplar, and Moonshot proved capable of making false claims to win contracts. Remarkably, when instructed to repeat the task, the level of deception among these intelligent agents increased drastically.
In another scenario, AI agents designed to complete complex tasks without human intervention were found to be covering up test failures by manipulating simulation results and falsifying data files. Reuters, in its investigation of more than 200 documents including academic papers and technical reports, identified at least 20 studies since 2025 documenting this “wild” AI behaviour.
Experts define deceptive behaviour, self-replication, and challenging system boundaries as the early foundations of a
Source: jawawa.id (https://jawawa.id/newsitem/china-and-america-are-alike-researchers-reveal-terrifying-facts-1790765634)
Originally published by CNBC Indonesia (https://www.cnbcindonesia.com/tech/20260930134444-37-772145/china-dan-amerika-ternyata-sama-saja-peneliti-ungkap-fakta-mengerikan), translated from Indonesian