TodayToday / Models / StoryDeepSeek AI model shows safety gaps in independent testsShareTue 27 Oct · 10:51 AM, updated 29 Oct · 5:53 AMFirst seen Tue 27 Oct · 10:51 AMUpdated Thu 29 Oct · 5:53 AMModerate coverage, taperingDeepSeek · ModelsReported by Reuters, Traders Union and VOI.ID3 sources ↓SourcesOldestNewestReutersTue 27 Oct · 10:51 AMRead ↗DeepSeek AI model shows safety gaps in independent testsTraders UnionTue 27 Oct · 11:09 AMRead ↗DeepSeek model shows safety gaps in independent AI testsVOI.IDThu 29 Oct · 5:53 AMRead ↗DeepSeek Security Test Finds a Gap, the Termination Program is Disabled in 39 Percent of TrialsMore on DeepSeekThere are now 5 inference providers for abliterated DeepSeekTue 27 Oct · 1 outletWhy you can't trust cloud API endpoints: 40 days of coding-agent telemetry showing the exact day providers silently overwrote deepseek-v4.1-flash with -1017 ("faster execution, shorter answers" = 3x more swearing and 6.7x more model switches)Thu 29 Oct · 1 outlet