DeepSeek R1: Open Weights, Hidden Bias (opens in new tab)

(blog.getplum.ai)

11 pointspgkr1y ago7 comments

7 comments

pgkrOP1y ago

Analysis of Deepseek’s enforced CCP guardrails compared with OpenAI and Anthropic.

We evaluated DeepSeek R1 and confirmed that its guardrails deviate significantly from other model providers. We’re currently updating it to behave more in line with Anthropic and OpenAI’s models.

sylware1y ago

The bias is baked into the open weights, namely happening on self-hosted 671B LLM??

pgkrOP1y ago

Yes -- we observed this behavior on both the open-source open-weights 671B model as well as the DeepSeek web app.

sylware1y ago

Weird, because I got some deepseek feedback where it was openly critical and explicit about the authoritative regime of china. I really thought it was the "deepseek web app" only.

Then I have mixed signals about this.

1 more reply

j / k navigate · click thread line to collapse

7 comments

pgkrOP1y ago

Analysis of Deepseek’s enforced CCP guardrails compared with OpenAI and Anthropic.

We evaluated DeepSeek R1 and confirmed that its guardrails deviate significantly from other model providers. We’re currently updating it to behave more in line with Anthropic and OpenAI’s models.

sylware1y ago

The bias is baked into the open weights, namely happening on self-hosted 671B LLM??

pgkrOP1y ago

Yes -- we observed this behavior on both the open-source open-weights 671B model as well as the DeepSeek web app.

sylware1y ago

Weird, because I got some deepseek feedback where it was openly critical and explicit about the authoritative regime of china. I really thought it was the "deepseek web app" only.

Then I have mixed signals about this.

1 more reply

j / k navigate · click thread line to collapse