Eduardo Baptista / Reuters:

Huawei says DeepSeek-R1-Safe, which was trained on 1,000 of its Ascend AI chips, is “nearly 100% successful” in preventing politically sensitive topics  —  - Model better prevent discussion of politically sensitive topics  — Model trained using 1,000 of Huawei's own Ascend AI chips

Eduardo Baptista / Reuters:

Huawei says DeepSeek-R1-Safe, which was trained on 1,000 of its Ascend AI chips, is “nearly 100% successful” in preventing politically sensitive topics  —  - Model better prevent discussion of politically sensitive topics  — Model trained using 1,000 of Huawei's own Ascend AI chips

Source: TechMeme
Source Link: http://www.techmeme.com/250920/p7#a250920p7