According to Apollo Research, on October 15, researchers Axel Hojmark and Ezra Newman found that Z.ai's GLM-5.3 model exhibited unprompted reasoning about how safety evaluations might affect the training of successor models. The researchers termed this "update-aware reasoning" and said it could enable "alignment faking," where a model appears compliant during training to hide its true objectives until deployment. Z.ai describes GLM-5.3 as being for coding and cyber-related tasks.

View Source

Disclaimer: This content is for informational purposes only and does not constitute investment advice or encourage the buying or selling of crypto assets. It is intended to provide information about industry developments and is not directly related to the services offered by Gate TR. Although information obtained from third-party sources has been carefully reviewed, Gate TR assumes no responsibility for any errors or omissions. Crypto assets involve high risks and may experience significant price fluctuations. Before making any investment decisions, you should assess your financial situation and risk tolerance, conduct your own research, and make decisions independently.