Perplexity lets you try DeepSeek R1 without the security risk, but it's still censored

Chinese startup DeepSeek AI and its open-source language models took over the news cycle this week. Besides being comparable to models like Anthropic's Claude and OpenAI's o1, the models have raised several concerns about data privacy, security, and Chinese-government-enforced censorship within their training. 

AI search platform Perplexity and AI assistant You.com have found a way around that, albeit with some limitations. 

Also: I tested DeepSeek's R1 and V3 coding skills - and we're not all doomed (yet) 

On Monday, Perplexity posted on X that it now hosts DeepSeek R1. The free plan gives users three Pro-level queries per day, which you could use with R1, but you'll need the $20 per month Pro plan to access it more than that. 

In another post, the company confirmed that it hosts DeepSeek "in US/EU data centers - your data never leaves Western servers," assuring users that their data would be safe if using the open-source models on Perplexity. 

"None of your data goes to China," Perplexity CEO Aravind Srinivas reiterated in a LinkedIn post. 

Also: Apple researchers reveal the secret sauce behind DeepSeek AI

DeepSeek's AI assistant, powered by both its V3 and R1 models, is accessible via browser or app -- but those require communication with the company's China-based servers, which creates a security risk. Users who download R1 and run it locally on their devices will still run into censorship of certain topics determined by the Chinese government, as it's built in by default. 

As part of offering R1, Perplexity claimed it removed at least some of the censorship built into the model. Srinivas posted a screenshot on X of query results that acknowledge the president of Taiwan. 

However, when I asked R1 about Tiananmen Square using Perplexity, the model refused to answer.

When I asked R1 if it is trained not to answer certain questions determined by the Chinese government, it responded that it's designed to "focus on factual information" and "avoid political commentary," and that its training "emphasizes neutrality in global affairs" and "cultural sensitivity."

"We have removed the censorship weights on the model, so it shouldn't behave this way," said a Perplexity representative responding to ZDNET's request for comment, adding that they were looking into the issue. 

Also: What to know about DeepSeek AI, from cost claims to data privacy

You.com offers both V3 and R1, similarly only through its Pro tier, which is $15 per month (discounted from the usual $20) and without any free queries. In addition to access to all the models You.com offers, the Pro plan comes with file uploads of up to 25MB per query, a 64k maximum context window, and access to research and custom agents. 

Bryan McCann, You.com cofounder and CTO, explained in an email to ZDNET that users can access R1 and V3 via the platform in three ways, all of which use "an unmodified, open source version of the DeepSeek models hosted entirely within the United States to ensure user privacy." 

"The first, default way is to use these models within the context of our proprietary trust layer. This gives the models access to public web sources, a bias towards citing those sources, and an inclination to respect those sources while generating responses," McCann continued. "The second way is for users to turn off access to public web sources within their source controls or by using the models as part of Custom Agents. This option allows users to explore the models' unique capabilities and behavior when not grounded in the public web. The third way is for users to test the limits of these models as part of a Custom Agent by adding their own instructions, files, and sources."

Also: The best open-source AI models: All your free-to-use options explained

McCann noted that You.com compared DeepSeek models' responses based on whether it had access to web sources. "We noticed that the models' responses differed on several political topics, sometimes refusing to answer on certain issues when public web sources were not included," he explains. "When our trust layer was enabled, encouraging citation of public web sources, the models' responses respected those sources, seemingly overriding prior political biases."