DeepSeek-R1 AI Model Said to Be Censoring China-Focused Prompts Raising Concerns Over Its Reliability
The firm created the dataset of prompts by seeding questions into a program and by extending it via synthetic data generation. The dataset was published in a Hugging Face listing as well on Google Sheets. Promptfoo stated that it was able to find 1,360 prompts, where most of them contain sensitive topics around China.
Axios on MSN3d
Open AI's Sam Altman calls China's DeepSeek's R1 "Impressive," but promises better modelsOpenAI CEO Sam Altman on Monday called Chinese AI startup DeepSeek's R1 an "impressive model" and pledged his company would ...
The upstart AI chip company Cerebras has started offering China’s market-shaking DeepSeek on its U.S. servers.
DeepSeek, the Chinese artificial intelligence startup that sent tech stocks reeling this week, sparked fresh concerns about U ...
Computer scientist and AI expert Andrew Ng didn't explicitly mention the significance of R1 being an open source model, but ...
A security report shows that DeepSeek R1 can generate more harmful content than other AI models without any jailbreaks.
DeepSeek R1's large language model collects a huge amount of user data and sends it to China. AI also distorts information ...
DeepSeek-R1, developed by a Chinese AI lab, is potentially highly competitive and shockingly cost-effective, and could be a ...
Are DeepSeek V3 and R1 the next big things in AI? How this Chinese open-source chatbot outperformed some big-name AIs in ...
DeepSeek’s success is not based on outperforming its U.S. counterparts, but on delivering similar results at significantly ...
Results that may be inaccessible to you are currently showing.
Hide inaccessible results
Feedback