Evaluating DeepSeek V4.1 Flash-1017 for loss of control and harmful manipulation risks

An independent third-party evaluation characterizes the systemic safety risks of the open-weight DeepSeek V4.1 Flash-1017 model, focusing on loss of control and harmful manipulation. The evaluation notes that while closed-source releases are routinely accompanied by comprehensive system cards, open-weight models receive less scrutiny. The work positions independent evaluation as critical as frontier language models grow increasingly capable and autonomous.

As frontier language models grow increasingly capable and autonomous, independent third-party evaluations are critical for characterizing their systemic safety risks. While closed-source releases are routinely accompanied by comprehensive system cards, open-weight models receive less scrutiny regard