The artificial intelligence (AI) industry is evolving rapidly, with companies like Deepseek pushing the boundaries of innovation. However, external factors—from government regulations to economic shifts—play a crucial role in shaping its growth and sustainability.
In this post, we’ll conduct a PESTLEC analysis to examine the political, economic, social, technological, legal, environmental, and competitive factors influencing Deepseek. Understanding these forces provides insight into the company’s opportunities and challenges in the dynamic AI market. Let’s explore how Deepseek is positioned in this ever-changing landscape.
Political factors
Government-backed: Deepseek is supported by the Chinese government due to its affiliation with the Chinese Academy of Information and Communications Technology (CAICT), a division of the Ministry of Industry and Information and Technology of China. According to (Gong et al., 2024), the Chinese government has prioritized AI governance as part of one of its national strategies since 2017.
International Relations: Deepseek has already been banned in some countries namely, the United States, South Korea, Australia, Italy and Taiwan thus far. Collectively these countries have echoed a sentiment that there are data privacy violations such as collecting email, phone numbers, passwords and date of birth and sharing the information with advertisers (Aljazeera, 2025).
Economic factors
Market expansion: Deepseek's meteoric rise has shaken up the AI market, giving established companies a run for their money. On January 27, when Deepseek was officially launched, Nvidia lost upwards of $593 billion of market value (Carew et al., 2025).
Cost efficiency: According to Deepseek they have bragged about how they have trained their model using $6 million using 2000 Nvidia GPUs against ChatGPT using $80 million - $100 million using 16000 H100 GPUs (Hanbury et al., 2025).
Social factors
Public trust: As discussed previously Deepseek has had its share of controversy regarding its user data privacy. According to South Korea's National Intelligence Service (NTS), "Unlike other generative AI services, it has been confirmed that chat records are transferable as it includes a function to collect keyboard input patterns that can identify individuals and communicate with Chinese companies servers such as volceapplog.com" (Yim, 2025).
User adoption: Since its launch Deepseek has surpassed ChatGPT in downloads from Apple's App Store and has had 2.6 million downloads since 28 January 2025 (Yim & Flensted 2025).
Technological factors
Innovation: As previously mentioned Deepseek spent less money to train their model making it more accessible for other businesses who wish to use it they have also found a way to run it more efficiently with a strategy called "Mixture-of-Experts," which is a method used to reduce the computing power needed and increase the quickness of providing answers that are of high quality (Hanbury et al., 2025).
Security vulnerabilities: According to Cisco they ran several tests on the Deepseek R1 model to see if it could withstand hacking exploits and their results showed that Deepseek R1 failed to block any cyber attacks when benchmarked against other AI models (Kassianik & Karbasi, 2025).
Legal factors
Data privacy regulations: Deepseek has already been flagged by international regulators regarding data privacy laws and multiple nations have begun their investigations into the matter of user data being sent to Chinese companies who back Deepseek (Chee, 2025).
Intellectual property concerns: OpenAI has accused Deepseek of AI model distillation which is when a teacher AI model transfers knowledge to a student AI model. This is done to keep the high-performing original model while minimizing its size and computational needs.
Furthermore, OpenAI is saying that Deepseek didn't grant permission for their IP and that they did this to make their AI model better by breaching computer fraud and abuse acts (Bennett & Robinson, 2025).
Environmental factors
Resource consumption: Currently US AI companies use Nvidia's H100 GPUs which are more powerful but require more power. Up to 100,000 of these could power a small city. On the other hand, Deepseek uses the H800 GPUs which are slightly weaker than the H100 however are reported to use half the energy of the H100, therefore putting less environmental strain and operational costs (Rinnovabili, 2025).
Sustainability initiatives: Although at the time of writing this Deepseek has not yet published an environmental report, however on initial thoughts Deepseek seem to be adopting a green AI model by optimizing energy consumption (Baniya, 2025).
Competitive factors
Market position: In my opinion, Deepseek has caused a market disruption hurting some of the big tech heavyweights by becoming one of the fastest-growing companies in the AI industry. However, this will motivate competitors to further innovate and develop better services going forward.
Security reputation: Due to security concerns this could provide an opportunity for competitors to rectify their vulnerabilities and turn them into a strength or at least stabilize them to not be exploited by hackers.
Deepseek’s rapid rise in the AI industry underscores both the opportunities and challenges that come with technological disruption. While its cost-efficient approach and innovative model architecture have shaken up the market, concerns around data privacy, security, and regulatory scrutiny could hinder its global expansion. Competitors will likely leverage these weaknesses to reinforce their positions, while governments intensify oversight of AI governance. Ultimately, Deepseek’s ability to navigate these external forces will determine whether it remains a disruptive force or faces setbacks in an increasingly competitive and regulated landscape.
References
- Al Jazeera. (2025, February 7). Which countries have banned DeepSeek and why? Al Jazeera. https://www.aljazeera.com/news/2025/2/6/which-countries-have-banned-deepseek-and-why
- Baniya, V. (2025, February 9). DeepSeek and Green Innovation: The Power of AI in Sustainability. Medium. https://medium.com/@stoooryboooard/deepseek-and-green-innovation-the-power-of-ai-in-sustainability-70a92143357d
- Bennett, M., & Robinson, R. (2025, February 3). OpenAI accuses DeepSeek of unlawful use of AI models, raising ethical and legal concerns. JD Supra. https://www.jdsupra.com/legalnews/openai-accuses-deepseek-of-unlawful-use-2896277/
- Carew, S., Cooper, A., & Banerjee, A. (2025, January 28). DeepSeek sparks AI stock selloff; Nvidia posts record market-cap loss. Reuters. https://www.reuters.com/technology/chinas-deepseek-sets-off-ai-market-rout-2025-01-27/
- Chee, F. (2025, February 11). DeepSeek may face further regulatory actions, EU privacy watchdog says. Reuters. https://www.reuters.com/technology/deepseek-may-face-further-regulatory-actions-eu-privacy-watchdog-says-2025-02-11/
- Flensted, T. (2025, January 28). DeepSeek AI Statistics and Facts (2025). SEO.AI. https://seo.ai/blog/deepseek-ai-statistics-and-facts
- Gong, J., Qu, H., & Dorwart, H. (2024, March 14). AI governance in China: strategies, initiatives, and key considerations. Bird & Bird. https://www.twobirds.com/en/insights/2024/china/ai-governance-in-china-strategies-initiatives-and-key-considerations#:~:text=At%20least%20since%202017%2C%20the,numerous%20planning%20documents%20and%20regulations.
- GSA. (2020, June 18). CAICT | GSA. https://gsacom.com/associate/caict/
- Hanbury, P., Wang, J., Brick, P., & Cannarsi, A. (2025, February 10). DeepSeek: a game changer in AI efficiency? Bain & Company. https://www.bain.com/insights/deepseek-a-game-changer-in-ai-efficiency/
- Hulse, Newby, T., Meyer, S., & Tsang, F. (2025, February 3). DeepSeek, model distillation, and the future of AI IP protection. Fenwick. https://www.fenwick.com/insights/publications/deepseek-model-distillation-and-the-future-of-ai-ip-protection
- Kassianik, P. (2025, January 31). Evaluating security risk in DeepSeek and other frontier reasoning models. Cisco Blogs. https://blogs.cisco.com/security/evaluating-security-risk-in-deepseek-and-other-frontier-reasoning-models
- Rinnovabili. (2025, January 29). How energy-efficient is DeepSeek, China’s AI disruptor? https://www.rinnovabili.net/business/markets/deepseeks-energy-consumption-ais-75-power-cut/
- Yim, H. (2025, February 11). South Korea spy agency says DeepSeek “excessively” collects personal data. Reuters. https://www.reuters.com/technology/artificial-intelligence/south-korea-spy-agency-says-deepseek-excessively-collects-personal-data-2025-02-10/?utm_source=chatgpt.com
- 守护AI安全,共建行业自律典范——首批17家企业签署《人工智能安全承诺》. (2024, December 24). Weixin Official Accounts Platform. https://mp.weixin.qq.com/s/s-XFKQCWhu0uye4opgb3Ng
0 Comments