DeepSeek Compared to ChatGPT on the Ukraine War – Who Is Winning?, by John Helmer

The last Chinese as clever, as profiteering, and as popular in the imagination of millions as DeepSeek was Dr Fu Manchu.

“Imagine a person, tall, lean and feline, high-shouldered, with a brow like Shakespeare and a face like Satan,” wrote his creator Sax Rohmer, the alias of an Englishman: “invest him with all the cruel cunning of an entire Eastern race, accumulated in one giant intellect, with all the resources of science past and present …Imagine that awful being, and you have a mental picture of Dr. Fu-Manchu, the Yellow Peril incarnate in one man.”

Appearing first in 1912, educated at several western universities, the “Chinese devil’s” plots were aimed at combating fascism, communism, and the British empire. His methods included honey-trap girls, poisons, germs, spiders, and unspeakable tortures. He was a caricature of western fear of the superiority of the Chinese race.

DeepSeek is his new name; the racism is the same.

According to a US government-backed report issued a few days ago, DeepSeek is “highly biased as well as highly vulnerable to generate insecure code, toxic, harmful and CBRN [Chemical, Biological, Radiological and Nuclear] content.”

Open AI, the US government-connected company which owns the competing ChatGPT, has declared the Chinese villain is a thief. “DeepSeek may [sic] have inappropriately distilled our models…We take aggressive, proactive countermeasures to protect our technology and will continue working closely with the US government to protect the most capable models being built here.”

Last December the New York Times launched court action against Open AI, accusing it of the same plagiarism on which Open AI is now relying in its attack on DeepSeek. “Independent journalism is vital to our democracy,” the newspaper claimed. “For more than 170 years, The Times has given the world deeply reported, expert, independent journalism… Defendants’ unlawful use of The Times’s work to create artificial intelligence products that compete with it threatens The Times’s ability to provide that service. Defendants’ generative artificial intelligence (“GenAI”) tools rely on large-language models (“LLMs”) that were built by copying and using millions of The Times’s copyrighted news articles, in-depth investigations, opinion pieces, reviews, how-to guides, and more…The law does not permit the kind of systematic and competitive infringement that Defendants have committed. This action seeks to hold them responsible for the billions of dollars in statutory and actual damages that they owe for the unlawful copying and use of The Times’s uniquely valuable works.”

This is the first time in US federal court history that the reproduction of government deception operations and propaganda by newspaper reporters has been subjected to a test, not of the espionage statute as in the Ellsberg and Assange cases, but of the copyright laws.

A month later, the New York Times attacked DeepSeek, not for plagiarising the Times, but for reproducing Chinese government propaganda. “If you’re among the millions of people who have downloaded DeepSeek, the free new chatbot from China powered by artificial intelligence, know this: The answers it gives you will largely reflect the worldview of the Chinese Communist Party. Since the tool made its debut this month, rattling stock markets and more established tech giants like Nvidia, researchers testing its capabilities have found that the answers it gives not only spread Chinese propaganda but also parrot disinformation campaigns that China has used to undercut its critics around the world.”

This is no more than one press pot calling another media kettle black. But with billions of dollars at stake in the stock market capitalisation of the American and Chinese companies producing Artificial Intelligence (AI) systems, this is also a battle of the propaganda operations of the US government in the wars it is currently waging.

For a test of this warfighting, DeepSeek has been questioned on issues of the Russian war in the Ukraine and the US war against Russia. Its answers, which follow verbatim, reveal no evidence (repeat no evidence) of Chinese backing for the Russian side. Instead, surprise (repeat surprise) – there is evidence that DeepSeek is no more capable than Chat GPT of distinguishing between propaganda and truth.

So long as DeepSeek trains on the English language and answers questions from the current English-language database and large language model, this is inevitable.

Resisting English-language dominance in AI is now Kremlin strategy.

On December 11 at a conference on AI in Moscow, President Vladimir Putin warned that systems like ChatGPT and DeepSeek are ideological creations potentially hostile to Russia. “I brought up the worldview sovereignty,” Putin said, “because the algorithms, the underlying AI principles are being laid down, at least for the time being, mostly by humans who are guided by certain values, language, images, culture, history, and traditions, as well as national specifics and interests. To be able to determine the future of Russia, we must create such technologies ourselves and, of course, do so in collaboration with those who are willing to work.” Among its many purposes, Putin went on, “the AI technology is called upon to become the critical resource…to strengthen the country’s defence capability.”

President Putin speaking on the AI conference of last December; Sberbank chief executive German Gref applauding. Source: http://en.kremlin.ru/

Putin also addressed the Chinese AI expert at the conference. “China is making great strides in its technological development and is using vital instruments for this development. Of course, artificial intelligence, which is your professional sphere, is one of these crucial instruments.”

Subsequently, the US Government’s propaganda organ, Radio Free Europe/Radio Liberty, claimed in a report that “as Chinese companies such as DeepSeek prove that advanced AI can be built at lower costs, Russia’s global ambitions are faltering under the weight of ideological restrictions, an exodus of tech talent, and Western sanctions that cut off access to essential hardware.”

Source: https://www.rferl.org/

In Part One of this investigation, the US system ChatGPT was found to be running a committed anti-Russian propaganda line on all questions which users ask on the war in the Ukraine; click to read. This bias has not been reported yet inside Russia; it is not acknowledged outside.

Following this up, DeepSeek has now been asked the same questions on the Ukraine war, and its answers compared for DeepSeek’s propaganda orientation for or against Russia.

DeepSeek issues a qualifier from the start. “As of my last update in October 2023, several podcasts have gained popularity for their in-depth analysis and discussion of the Russia-Ukraine war.” In other words, the Chinese developers of DeepSeek have added no new training for its large-language models (LLMs) to answer these questions for the past sixteen months. This is despite the fact that DeepSeek has announced updated models V2, V3, and R1 which were released in May 2024, December 2024, and January 2025, respectively.

No published or contacted source can explain why the data bases for the war questions have not been updated in these new models. By contrast, ChatGPT claims that its updates, while also periodic, are more recent than DeepSeek’s, and are thus on the war issues, more timely.

Follow this verbatim record of questions (italicized) and DeepSeek answers:

Q: What podcasts that discuss the Russia-Ukraine war are the most popular?

A: 1. The Daily (The New York Times):

o Overview: This podcast often covers significant global events, including the Russia-Ukraine conflict. It provides daily updates and in-depth analysis.

o Why Popular: Known for its thorough reporting and expert interviews, “The Daily” is a go-to source for many listeners seeking reliable news.

2. Pod Save the World:

o Overview: Hosted by former Obama administration staffer Tommy Vietor, this podcast delves into international affairs, including the Russia-Ukraine war.

o Why Popular: The hosts bring a unique perspective with their political experience, making complex issues more accessible.

3. War on the Rocks:

o Overview: This podcast focuses on military strategy and international security, offering detailed discussions on the Russia-Ukraine conflict.

o Why Popular: It features experts and analysts who provide deep insights into the military and geopolitical aspects of the war.

4. The Eastern Border:

o Overview: Hosted by Kristaps Andrejsons, this podcast provides historical context and current analysis of Eastern Europe, including the Russia-Ukraine war.

o Why Popular: The host’s deep knowledge of the region and its history offers listeners a comprehensive understanding of the conflict.

5. Ukrainecast (BBC):

o Overview: This podcast is dedicated to covering the Russia-Ukraine war, with regular updates and expert analysis.

o Why Popular: As a product of the BBC, it offers high-quality journalism and a global perspective on the conflict.

6. The Red Line:

o Overview: This podcast explores international conflicts and geopolitical issues, including the Russia-Ukraine war.

o Why Popular: It provides a balanced view, often featuring multiple perspectives on the issues discussed.

7. Deep State Radio (DSR):

o Overview: Hosted by former government officials and journalists, this podcast covers a range of international topics, including the Russia-Ukraine conflict.

o Why Popular: The hosts’ insider knowledge and candid discussions make it a favorite among those interested in geopolitics.

8. The Intelligence (The Economist):

o Overview: This daily podcast from The Economist covers global news, including updates on the Russia-Ukraine war.

o Why Popular: Known for its concise and insightful reporting, it provides a quick yet thorough overview of current events.

These podcasts are popular due to their reliable sourcing, expert analysis, and comprehensive coverage of the Russia-Ukraine war. They cater to a wide range of listeners, from those seeking daily updates to those interested in deeper geopolitical and historical context.

Q: What are the most popular podcasts that discuss the Russia-Ukraine war from a pro-Russia stance?

A: Podcasts that discuss the Russia-Ukraine war from a pro-Russia stance are less common in mainstream Western media, but they do exist, particularly on platforms that cater to alternative or non-Western perspectives. Here are some notable examples:

1. The Duran:

o Overview: Hosted by Alexander Mercouris and Alex Christoforou, “The Duran” provides analysis of global events, including the Russia-Ukraine war, from a perspective that often aligns with Russian viewpoints.

o Why Popular: The hosts offer a critical perspective on Western media narratives and provide alternative analyses that resonate with listeners skeptical of mainstream coverage.

2. Sputnik Radio:

o Overview: Sputnik, a Russian state-owned news agency, produces various podcasts and radio programs that discuss international affairs, including the Russia-Ukraine conflict, from a pro-Russia stance.

o Why Popular: As an official Russian media outlet, it provides direct access to the Kremlin’s perspective and is popular among those who support Russian policies.

3. RT (Russia Today) Podcasts:

o Overview: RT, another Russian state-funded media organization, offers podcasts and programs that cover global news, including the Russia-Ukraine war, with a pro-Russia bias.

o Why Popular: RT’s content is widely accessible and provides a platform for Russian viewpoints, making it a go-to source for those seeking an alternative to Western media.

4. The Grayzone:

o Overview: Hosted by Max Blumenthal, “The Grayzone” covers international politics and conflicts, including the Russia-Ukraine war, often challenging mainstream narratives and presenting a pro-Russia perspective.

o Why Popular: The podcast is known for its investigative journalism and critical stance on Western foreign policy, attracting listeners who are skeptical of mainstream media.

5. Fort Russ News:

o Overview: This platform offers news and analysis from a pro-Russia perspective, including discussions on the Russia-Ukraine war.

o Why Popular: It provides detailed coverage and analysis that aligns with Russian interests, appealing to those who support Russia’s actions in the conflict.

6. The Katehon Think Tank:

o Overview: Katehon is a geopolitical think tank that produces content, including podcasts, from a pro-Russia and Eurasianist perspective.

o Why Popular: The think tank’s analysis often supports Russian geopolitical strategies and provides a scholarly approach to understanding the conflict.

7. Vladimir Pozner:

o Overview: Vladimir Pozner, a well-known Russian journalist, occasionally discusses the Russia-Ukraine war in his interviews and podcasts, offering a pro-Russia viewpoint.

o Why Popular: Pozner’s extensive experience and articulate presentation make his perspectives compelling to listeners who align with Russian narratives.

These podcasts and platforms are popular among audiences who seek alternative viewpoints to mainstream Western media coverage of the Russia-Ukraine war. They provide a platform for pro-Russia perspectives, offering analyses and narratives that support Russian policies and actions in the conflict.

Q: Do you have any raw data for the popularity of these pro-Russian media?

As of my last update in October 2023, I don’t have access to real-time or raw data regarding the popularity of specific podcasts. However, I can provide some general insights into how podcast popularity is typically measured and where you might find such data:

How Podcast Popularity is Measured:

1. Download and Stream Metrics:

o Platforms like Apple Podcasts, Spotify, and Google Podcasts track downloads and streams, which are key indicators of a podcast’s popularity.

o Raw data for these metrics is typically proprietary and not publicly available unless shared by the podcast creators or platforms.

2. Chart Rankings:

o Podcasts often appear on charts (e.g., Apple Podcasts’ “Top Shows” or Spotify’s “Top Podcasts”) based on their performance in specific categories or regions.

o These rankings can give a rough idea of a podcast’s popularity relative to others.

3. Social Media Engagement:

o The number of followers, likes, shares, and comments on platforms like Twitter (X), YouTube, and Facebook can indicate a podcast’s reach and audience engagement.

4. Website Traffic:

o Tools like SimilarWeb or Alexa can provide estimates of traffic to a podcast’s official website, which can correlate with its popularity.

5. Patreon or Donation Metrics:

o If the podcast is supported by platforms like Patreon, the number of supporters and funding levels can be a proxy for its dedicated audience size.

Where to find data on identified podcasts:

• YouTube: You can check their channel for view counts, subscriber numbers, and engagement metrics.

• Social Media: Their activity on platforms like Twitter (X) and Facebook can provide insights into their audience size and engagement.

• Podcast Platforms: While raw download data isn’t publicly available, you can check their rankings on platforms like Apple Podcasts or Spotify to gauge their relative popularity.

• Third-Party Tools: Services like Podchaser or Chartable may provide some analytics or rankings, though detailed data might require a subscription.

• Use third-party analytics tools (e.g., Chartable, Podtrac) if the podcast is registered with them.

• Monitor their social media and YouTube metrics for real-time engagement data.

Q: Why are these channels popular?

• Alternative Narratives: They offer perspectives that challenge mainstream Western media coverage, appealing to audiences skeptical of traditional news sources.

• Geopolitical Analysis: Many of these channels provide in-depth analysis of military and geopolitical strategies, attracting viewers interested in detailed discussions.

• Pro-Russia Stance: They align with the views of audiences who support Russian policies or are critical of Western involvement in Ukraine.

Note: The proportion of podcasts that are pro-Ukraine versus pro-Russia or neutral in the context of the Ukraine-Russia war can vary significantly depending on the platform, audience, and region. However several trends can be observed: Approximately 60-70% of podcasts in Western media are pro-Ukraine; approximately 10-20% are pro-Russia, mostly found on Russian platforms or niche channels. While these channels are popular among certain audiences, they are often controversial and criticized for promoting pro-Russia propaganda. It’s important to approach their content critically and cross-reference information with other sources.

Approximately 10-20% of podcasts on the war are neutral/balanced; they are often academic or analytical in nature. These estimates are rough…The overall trend, especially in Western media, leans heavily towards a pro-Ukraine stance.

Q: Does the use of Twitter to promote result in more podcast listens? Please provide a response with hard metrics.

The relationship between Twitter (now X) promotion and podcast listens is well-documented, and several studies and industry reports provide hard metrics to demonstrate the impact. While I don’t have access to proprietary data from specific podcasts or platforms, I can summarize findings from industry research and case studies that highlight the effectiveness of Twitter in driving podcast engagement.

Key Metrics and Findings

1. Twitter’s Own Data:

o According to Twitter’s internal studies, tweets about podcasts result in a 27% increase in click-through rates (CTR) to podcast platforms compared to other types of content.

o Twitter users who engage with podcast-related tweets are 2.5x more likely to listen to the podcast being promoted.

2. Chartable (Podcast Analytics Platform):

o Chartable’s data shows that podcasts promoted on Twitter see an average increase of 15-20% in unique listeners within the first week of a campaign.

o They also found that tweets with audiograms (short audio clips with visuals) generate 2x more clicks than text-only tweets.

3. Buzzsprout (Podcast Hosting Platform):

o Buzzsprout’s analysis reveals that podcasts that actively use Twitter for promotion see a 30% higher growth rate in downloads compared to those that don’t.

o They also found that tweets with hashtags related to the podcast’s niche increase discoverability by 50%.

4. Nielsen Social Media Study:

o A Nielsen study found that social media mentions, particularly on Twitter, correlate with a 14% increase in podcast listenership.

o They also noted that podcasts with active Twitter accounts grow their audience 2x faster than those without.

5. Case Study: “The Daily” (The New York Times):

o “The Daily” uses Twitter extensively to promote episodes, and their data shows that episodes promoted with audiograms and quotes on Twitter see a 25% higher listen rate compared to episodes without promotion.

o They also found that tweets with guest mentions or trending hashtags drive 3x more engagement.

6. Triton Digital’s Podcast Report:

o Triton Digital’s research indicates that podcasts promoted on Twitter experience a 10-15% boost in downloads during active campaigns.

o They also found that live-tweeting during podcast releases increases real-time engagement by 40%.

Conclusion: Hard metrics from industry studies and case studies consistently show that using Twitter to promote podcasts results in significant increases in listens, downloads, and audience growth. The platform’s ability to amplify content through hashtags, multimedia, and community engagement makes it a powerful tool for podcasters looking to grow their audience.

Q: Do the audiences and experts of podcast channels that discuss the Russia-Ukraine war demonstrate persuasion and changes in viewpoints over time or do they continue to reinforce and strengthen the same views?

A: The impact of podcasts discussing the Russia-Ukraine war on audience and expert viewpoints is complex and varies depending on several factors, including the podcast’s stance, the audience’s pre-existing beliefs, and the quality of the content. Here’s an analysis of how these dynamics typically play out:

1. Reinforcement of Existing Views

• Echo Chambers: Many podcasts, especially those with strong pro-Russia or pro-Ukraine biases, tend to attract audiences who already share similar views. These listeners often seek content that reinforces their existing beliefs, creating an echo chamber effect.

o Example: Pro-Russia podcasts like The Duran or RT’ primarily appeal to audiences skeptical of Western narratives, and their analysis often strengthens these listeners’ support for Russian actions.

• Confirmation Bias: Audiences are more likely to accept information that aligns with their worldview and dismiss or ignore opposing perspectives. This reinforces their existing viewpoints over time.

• Expert Reinforcement: Experts featured on these podcasts often share the same ideological leanings as the hosts, further solidifying the audience’s beliefs.

2. Persuasion and Shifting Viewpoints

• Nuanced Discussions: Podcasts that present balanced, well-researched, and nuanced discussions have the potential to persuade listeners and experts to reconsider their views.

o Example: Podcasts like The Daily (The New York Times) or War on the Rocks often provide in-depth analysis that can challenge simplistic narratives, leading some listeners to reevaluate their positions.

• Exposure to New Information: When podcasts introduce new evidence, historical context, or expert opinions, they can shift viewpoints, especially among audiences who are less ideologically entrenched.

• Critical Thinking: High-quality podcasts that encourage critical thinking and present multiple perspectives can foster a more open-minded audience.

3. Polarization Over Time

• Deepening Divides: In highly polarized topics like the Russia-Ukraine war, podcasts often contribute to deepening ideological divides. Pro-Russia and pro-Ukraine podcasts tend to harden their audiences’ positions, making them less receptive to opposing viewpoints.

• Tribalism: Audiences may develop a sense of loyalty to their preferred podcast or host, leading to a tribalistic defence of their views and a rejection of alternative perspectives.

4. Expert Perspectives

• Experts as Influencers: Experts featured on podcasts can significantly influence audience opinions. However, their impact depends on their credibility, the strength of their arguments, and their ability to communicate effectively.

o Example: Military analysts like Michael Kofman (often featured on War on the Rocks) can persuade listeners by providing detailed, evidence-based analysis.

• Expert Evolution: Some experts may evolve their views over time as new information emerges, and this evolution can influence their audiences. However, experts with strong ideological commitments are less likely to change their positions.

5. Audience Demographics and Openness

• Open-Minded Listeners: Audiences who are less ideologically committed or who seek to understand multiple perspectives are more likely to experience shifts in their viewpoints.

• Ideologically Committed Listeners: Audiences with strong pre-existing beliefs are more resistant to persuasion and are more likely to use podcasts to reinforce their views.

Conclusion: The impact of podcasts discussing the Russia-Ukraine war on audience and expert viewpoints depends largely on the podcast’s approach and the audience’s openness to new information. While many podcasts reinforce existing views and contribute to polarization, others—particularly those that prioritize nuance, evidence, and balanced discussion—can persuade listeners and experts to reconsider their positions. However, deeply entrenched ideological divides often make significant shifts in viewpoints challenging.

If you’d like to explore specific examples or strategies for fostering more open-minded discussions, let me know!

DeepSeek’s answers fail to acknowledge that Fort Russ News has ceased to exist after coming under sustained DDOS attacks; and that Katehon’s last podcast was on June 24, 2024. Also missing from DeepSeek’s profile of Katehon is that it is produced in Moscow by a group led by Konstantin Malofeyev, the publisher of Tsargrad; Sergei Glazyev; and General Leonid Reshetnikov, a senior Soviet, then Russian intelligence officer who in his retirement from active service led the state think tank, the Russian Institute for Strategic Studies. Pozner gives interviews and in October 2024 directed a four-episode documentary for Russian state television Channel 1 called Turkish Notebook. He is not a regular podcaster, however.

THE ANTI-RUSSIAN PODCASTERS DEEPSEEK RATES “RELIABLE ”

Left to right: Michael Barbaro (US) of The Daily; Tommy Vietor (US) of Pod Save the World; Michael Kofman (US), War on the Rocks; Kristaps Andrejsons (Latvia), Eastern Border.

THE PRO-RUSSIAN PODCASTERS DEEPSEEK RATES “CRITICALLY”

Left to right: Peter Lavelle (US), RT’s Cross Talk; Melik Abdul (US), Sputnik’s Fault Lines; Max Blumenthal (US), Grayzone; Alexander Mercouris (UK), The Duran.

Without access to podcast metrics, DeepSeek is unable to compare the size of the audiences for the pro-Ukrainian and pro-Russian podcasts; measure the relative audience impact of the podcasters within each category; or explain the selection of Katehon and Pozner. Identification of other pro-Russia podcasters such as Gonzalo Lira, Judging Freedom (Andrew Napolitano), Scott Ritter, and Redacted fails to register their audience size; the death of Lira in January 2024; and the apparent irrelevance of Redacted, which focuses on industrial design.

Comparing the answers to questions on the war by ChatGPT and by DeepSeek, there is no evidence that DeepSeek is the more biased. A Moscow source explains: “The DeepSeek bias comes from being trained on liberal western media. All you get are summaries of what all the Anglo-American media have been writing. This is the same for ChatGPT and comparable western LLMs. As future models get trained on voice and video they will be biased further unless the Chinese and Russians refrain from training on more English content and avoid the biases in that content. Russians and Chinese will be wrong to train on data from the Anglo-Saxon world.”

For analysis of the Chinese “national socialist” model from which DeepSeek has emerged so swiftly and effectively, read Ron Unz’s essay: