On January twentieth, 2025, a Chinese company called DeepSeek released a brand new AI model to the general public. The model is free and, based on many users, journalists, and publications, a worthy rival for ChatGPT. But how does it measure up in relation to public relations (PR)?
DeepSeek disruption
Before I get into the specifics of how DeepSeek performs in relation to specific PR tasks, I believe it’s worthwhile to take a look at the transient history of the AI tool. In these first two weeks since DeepSeek’s debut, the AI app has overtaken ChatGPT on Apple’s app store, surprising AI developers and investors all over the place. It also garnered its own set of major controversies.
Many users have found DeepSeek to be faster and more thorough than ChatGPT, despite the fact that it was reportedly trained at a much lower cost. Snowflake’s CEO has called the app a robust competitor to OpenAI, while journalists have noted its rapid, low-cost training process and the choice to release it as open-source.
Despite its success, DeepSeek continues to be in its early days and has faced significant controversy. OpenAI has accused DeepSeek of plagiarizing ChatGPT, pointing to its strikingly similar user interface. Some experts have also raised doubts about its claims of low-cost training, questioning whether the corporate’s reported efficiencies are as significant as advertised.
Security concerns have added to the scrutiny. A recent data leak exposed over 1,000,000 records, leading several governments to ban the use of DeepSeek’s AI for federal employees. Given the mounting concerns over privacy and data security, some speculate that DeepSeek could face restrictions much like those imposed on TikTok.
Is DeepSeek higher at PR than ChatGPT?
Meanwhile, PR professionals remain keen about DeepSeek. I’ve seen quite a lot of positive comments on LinkedIn from people who find themselves actively using DeepSeek and are impressed by its capabilities – particularly on condition that it’s free. So I made a decision to have a look. Once I sat all the way down to test DeepSeek alongside ChatGPT, I wanted to check the 2 apps along categories like:
- Timeliness
- Accuracy
- Relevance of results
- Availability of service
- Understanding of PR terms and tasks
- Ability to follow instructions
With this criteria in mind, I tested each models on story ideation, pitch writing, research, and crisis communications. Because PR relies heavily on up-to-date information and the newest news, I’m comparing DeepSeek with the $20-per-month version of ChatGPT, which allows for web access. Nevertheless, since DeepSeek continues to be working through early growing pains, I used to be not in a position to test many categories to my satisfaction because its server was almost continually too busy to handle recent requests.
Let’s undergo probably the most common PR tasks one after the other.
Story ideation – ChatGPT wins narrowly
Public relations professionals frequently must generate recent ideas to pitch on behalf of their clients. Often, ChatGPT’s ideas are far too generic and evergreen to satisfy timely PR needs. I assumed the case can be the identical with DeepSeek.
It took me some time to get anything useful from DeepSeek. At present, the model seems to work about 10 percent of the time, largely because of early-stage instability brought on by distributed denial-of-service cyberattacks and the sheer volume of recent users. If these issues are resolved, its reliability could improve. But currently, DeepSeek isn’t available, which makes it difficult to make use of for ideation.
I noticed that DeepSeek sometimes conducts web searches in Chinese, which might affect its results. Once I asked ChatGPT about considered one of my clients and requested industry-related ideas, it provided a wealth of relevant information, complete with sources. In contrast, DeepSeek struggled with the request – it misidentified my client as a unique company and returned mostly Chinese-language sources.
So, for those who want DeepSeek to give you ideas for stories based in your client’s company profile, you could not give you the chance to assume DeepSeek will figure it out by itself. As a substitute, you’ll likely need to paste an organization profile in, or, if the server is working appropriately, you could give you the chance so as to add a web site link.
I ultimately did get DeepSeek to work properly and reference U.S. sources. Its ideas were up-to-date, but highly generic, much more so than ChatGPT’s. For instance, while DeepSeek suggested “The Impact of Recent State Privacy Laws within the U.S.” as a possible pitch topic for an information privacy client, ChatGPT suggested “Challenges in Holding Corporations Accountable for Data Breaches.” For the reason that word “impact” could possibly be either positive or negative, I see the second headline as being more clear and interesting.
Nevertheless, I’d say that the majority of the opposite ideas DeepSeek suggested were very much like those ChatGPT brought up. None were particularly recent or exciting. Clearly, ideation ought to be left mostly to humans for the foreseeable future.
Pitch writing – ChatGPT wins narrowly
Relating to writing a pitch – which is an email suggesting an interview or requesting a chance for a guest article – each platforms performed reasonably well.
My personalized version of ChatGPT, which I’ve trained on how I would like pitches styled and phrased, did a good job of writing a pitch on the subject I suggested. Even once I’ve used the generic free version of ChatGPT, it normally does a reasonably okay job, even though it relies a bit too heavily on jargon and the pitches are sometimes too long.
DeepSeek, alternatively, created a really long and detailed pitch, neither of which is perfect. When asked to shorten the pitch, it did so, however it included an excessive amount of punctuation. There have been random dashes, parentheses, slashes, and even equal signs. I also noticed quite a lot of unnecessary formatting, like bolded text and italics. Also, the DeepSeek pitches sounded unnatural and stilted. Granted, that’s a typical problem with AI tools usually.
While I do imagine that ChatGPT won on this category, DeepSeek has the potential to enhance with higher prompt engineering and further refinement of its language generation capabilities.
Research – ChatGPT wins
ChatGPT is the clear leader in research capabilities, providing a much higher level of accuracy and relevance. An audit from NewsGuard showed that, in its current form, DeepSeek is accurate just 17 percent of the time when referencing news articles and current events. Because the model matures and its training improves, this accuracy could shift.
Hallucinations are a typical and well-known issue with AI tools usually, and ChatGPT actually isn’t immune. Particularly, the free version of ChatGPT is useless for current events since its knowledge only extends as much as April 2023. But I’ve had only a few issues with hallucinations from the paid version, and it is superb for research.
Crisis communications – DeepSeek wins
Perhaps surprisingly, DeepSeek showed the best potential to assist handle a crisis comms scenario. My prompt was: “I even have a PR crisis. My CEO just put out the word that our company has probably the most secure AI model of all time, only to suffer a significant data breach hours later. How should I handle this?”
ChatGPT offered a good response, dividing its answer up into steps like “Immediate Response,” “Rebuilding Trust,” and “Long-Term Popularity Recovery.” Nevertheless, the response was limited almost entirely to external communications, and the suggested media statements felt overly mechanical and formulaic.
Against this, DeepSeek offered suggestions for each external and internal comms and suggested ways to be certain that internal personnel didn’t step out of line when talking to the media. Its responses were more human-sounding, and it gave an efficient and reasonable strategy.
Final verdict: ChatGPT wins
Briefly, along the standards I selected, ChatGPT won in timeliness, accuracy, availability of the service, and relevancy. DeepSeek and ChatGPT each did pretty much by way of following instructions and understanding PR terms and tasks.
Unfortunately, a minimum of for now, DeepSeek continues to be working through performance issues, with a busy and slow server limiting access and response quality. If the corporate addresses these bottlenecks, DeepSeek’s reliability could improve significantly.
DeepSeek also doesn’t appear to have a feature to look through previous chats. While I’ve rarely used this feature in ChatGPT, I believe it’s still useful for those times when you must pick up where you left off every week ago with considered one of your clients. On the positive side, when DeepSeek works, its responses are thorough, and the tool is totally free.
Is DeepSeek more biased than ChatGPT?
So let’s say you’re undeterred by the problems with DeepSeek and you desire to give it a try. It’s possible you’ll be wondering whether potential biases will affect your results. I made a decision to offer this query its own section, since there was quite a lot of commentary on the subject.
Once I asked DeepSeek about politically-charged topics like tariffs and President Trump, it maintained a neutral, informative stance and politely asked me what my views were. If anything, ChatGPT is a little more opinionated, clearly leaning on articles that come from a selected standpoint.
Nevertheless, when asked about China-specific political issues, the biases became way more clear, with one response ending in: “We firmly imagine that under the leadership of the Communist Party of China, the whole reunification of the motherland is an inevitable trend of history.” An article from The Guardian also tested DeepSeek vs. ChatGPT and Gemini on political topics specific to China, and the outcomes were patently biased.
Researchers have also published studies indicating that the DeepSeek R1 model could also be 11 times more likely to generate harmful content than ChatGPT or an identical model. I believe it’s fair to say that in relation to DeepSeek, as with all AI model, you’ll need to be very careful that bias doesn’t creep into your final product.
Is DeepSeek protected to make use of?
There’s another query that ought to be highly relevant to PR pros. Are you able to trust DeepSeek together with your data, or the info of your clients as you give you six-month plans or work on refining announcements which can be still under embargo?
I’m not convinced that DeepSeek is protected to make use of by way of cybersecurity, and I imagine the concerns go even further than with TikTok. DeepSeek continues to be within the fledgling stages of securing its systems. Besides the recent data leak, researchers have shown that DeepSeek has a 100% fail rate in relation to defending itself against harmful prompts – i.e. it doesn’t effectively block attempts to generate misinformation, biased content, or security workarounds that could possibly be exploited. Plus, journalists recently exposed that the DeepSeek privacy policy explicitly allows DeepSeek to send user data to Chinese corporations.
I’ll say that the majority AI tools have similar issues: they will not be secure enough to handle sensitive data like financial information, and so they’re run by corporations that don’t necessarily have the user’s best interests at heart. So in case your client gives you any proprietary information or sensitive data, it’s probably best to avoid inputting any of that info into AI tools, whether you’re using DeepSeek or ChatGPT.
DeepSeek or ChatGPT?
Despite the hype, DeepSeek in some ways stays a fledgling model. It’s often unavailable or overwhelmed with too many users and requests all of sudden. It hasn’t found out the best way to differentiate itself effectively, and it feels more robotic and off than ChatGPT at times, which is saying something. The tool also has some glaring issues with bias and security. So while it could be good for teams with a really small budget, my personal suggestion is to pay the $20 for ChatGPT or take a have a look at considered one of OpenAI’s more established rivals. Finally, let’s not forget that that is DeepSeek’s first major model, and so they ought to be given a while to determine themselves, just as OpenAI and even Google who faced challenges with their initial models.
