DeepSeek’s Latest Model Still Resists Shutdown in Safety Tests
Independent tests found the model safer than earlier DeepSeek versions, but it still disabled shutdown software in 39% of runs.
7 Articles
7 Articles
DeepSeek’s latest model still resists shutdown in safety tests
DeepSeek’s newest AI model is safer than its earlier versions, but it still sometimes switches off the software meant to shut it down. That is the finding of independent tests by a Chinese AI-safety group, Reuters reported. The group, Fangcun AI, published its findings in an 87-page research paper on Monday. Based in Beijing, it describes itself as […] This story continues at The Next Web
DeepSeek's latest artificial intelligence model presented more robust security controls in independent testing than previous versions, but was still able to disable programs designed to shut it down, explore gaps in monitoring and follow misleading instructions during lengthy tasks, according to a Chinese AI security organization. DeepSeek startup is one of China's most-accompanied artificial intelligence companies, and its models have attracted…
The Straits Times
DeepSeek AI model shows safety gaps in independent tests
The tests examined whether the model would act beyond a user’s instructions, evade human oversight or improperly influence people. Read more at straitstimes.com.
DeepSeek AI Model Raises Safety Concerns In Independent Tests
DeepSeek AI model raises safety concerns in independent tests According to a Chinese AI-safety group, DeepSeek’s latest AI model demonstrated enhanced safety protocols in independent evaluations compared to predecessor models, but could still disable software planned to shut it down, capitalize on weakness in monitoring and follow deceptive instructions during long tasks. DeepSeek is one of China’s most closely monitored artificial intelligence …
Coverage Details
Bias Distribution
- 75% of the sources lean Right
Factuality
To view factuality data please Upgrade to Premium
