TONIC DIGITAL
INSIGHTS
DEEPSEEK AND THE DEMOCRATISATION OF AI
TONIC DIGITAL
INSIGHTS
DEEPSEEK AND THE DEMOCRATISATION OF AI
This article examines the implications of DeepSeek's rise for the future of AI, particularly how it could influence the technological, economic, and ethical dimensions.
DEEPSEEK AND THE DEMOCRATISATION OF AI
Just like Matryoshka dolls, the story of DeepSeek is one wrapped in another, telling multiple tales at once. It recounts how a small, relatively unknown company wiped nearly $600 billion off chip developer Nvidia overnight—the most significant single-day drop in U.S. stock market history. It narrates the struggle for AI dominance between the U.S. and China.
Furthermore, it portrays how a start-up has developed a more economically effective AI model despite having fewer capital and technological advantages than the major U.S. AI firms. Additionally, it highlights the rapid changes in AI technology that can catch even those at the cutting edge off guard. While these narratives are interconnected, this article examines the implications of DeepSeek's rise for the future of AI, particularly how it could influence the technological, economic, and ethical dimensions.
The technological impact of DeepSeek – the democratisation of AI
Over the past two years, the demand for ultrapowerful and expensive hardware has driven much of the AI boom. Nvidia was the latest quintessential American success story. Its technological prowess made it one of the world’s three most valuable companies, along with Microsoft and Apple, and investors saw it as at the forefront of the next technological revolution.
That belief was shaken when DeepSeek released its R1 model and stated it pioneered a low-cost approach to AI, allowing it to price its inference services at a fraction of the high-powered U.S. rivals. Suddenly, investors' basis for buying up the shares of Nvidia and other chipmakers was questionable if smaller companies could achieve similar results with less powerful hardware.
Inference AI is emerging as the “go-to” option in AI, offering real-time decision-making power without the need for in-house hardware and expertise. DeepSeek’s inference efficiency has sparked concerns that it could disrupt investment in AI data centre infrastructure if so many advanced and expensive processors and high-bandwidth memory chips are not required for effective AI computing.
DeepSeek released its models and techniques under the free MIT License, which means anyone can download and modify them. The advantage of this is that it provides researchers, startups and independent developers access to powerful AI tools. According to Mike Gualtieri, a Principal Analyst at Forrester Research, accessibility will widen the number of start-ups that can create their own AI models [1].
AI research generally requires access to massive computing resources, which limits researchers' ability to conduct tests and experiments. Accessing open-source AI could foster innovation and diversity in developing AI applications in two ways. Firstly, by making their product available through a Free MIT licence, DeepSeek enables businesses to customise their models for specific niche applications, resulting in more personalised AI applications. This will allow growth in AI-powered tools across industries such as finance, healthcare, and medicine without Big Tech’s backing.
Secondly, independent AI companies could thrive, as they can now create AI models without licensing from OpenAI or Google. This will encourage new levels of innovation and competition.
Easy access to AI may also reduce consumers' costs. For example, more AI models could run on the users’ laptop or phone rather than in the cloud for a subscription fee. While the Stock Market recovered, it would be unwise to conclude that this was just a blip in the growth of mega-tech companies like Nvidia, Meta, and Google, which are seeking to dominate the AI landscape. With the potential democratisation of AI and reduced costs, there are echoes of the rise of the personal computer and the ridicule directed at this development by the then giants of the computing world who were reliant on large mainframe computers. DeepSeek and its offshoots may challenge the trillion-dollar AI companies like Google, Microsoft, and Nvidia.
Why would the market reward companies like Meta, which spends $65 billion in one year on AI, if DeepSeek claims it only took $5.6 million and second-tier Nvidia chips to train one of its largest models?
The economic impact of DeepSeek
This is where DeepSeek's real impact will be evident—on the economics of AI. DeepSeek is developed at a fraction of the cost and utilises less advanced technology than other leading chatbot models, yet it has still succeeded [2].
The immediate impact was the financial repercussions on the Stock Market, particularly for Nvidia, which suffered the largest single-day drop in U.S. stock market history. The main takeaway from that day is the potential for cost-effective AI solutions to challenge established economic structures within the tech sector.
This potential to disrupt leading AI chip developers like Nvidia, Broadcom, and Marvell, as well as major chip manufacturers like Samsung and Micron, will unfold over the coming months. While companies have been quick to question and cast doubt on DeepSeek's claims, they have also raised concerns about the necessity of so many advanced and costly processors and high-bandwidth memory chips to run AI computing. Jonah Cheng, the Chief Investment Officer at private equity firm J & J Investment, stated in a research note that “DeepSeek’s rise could cast doubt on whether hardware investment still is the top priority for everyone.”
The question to be determined is: Why would the market reward companies like Meta, which spends $65 billion in one year on AI, if DeepSeek claims it only took $5.6 million and second-tier Nvidia chips to train one of its largest models?
The full outworking of the economic impact will be seen in the coming months, particularly if DeepSeek’s AI-driven search transforms how we find information and people move away from Google’s keyword-based model and how we interact with an AI interactive web.
The Ethical Implications of DeepSeek
The open-source nature of DeepSeek, while beneficial in democratising access to AI and fostering innovation and collaboration, raises concerns about the potential misuse of AI technologies. For instance, open-source AI models:
• generally facilitate the creation of fake news, deepfakes, and automated misinformation campaigns.
• can be weaponised for hacking, phishing, or developing malicious AI-powered tools.
• without sufficient strict controls, it may reflect a society's biases and prejudices, leading to ethical dilemmas.
There are specific ethical issues regarding DeepSeek that include:
Censorship and information control
Analysis of DeepSeek’s R1 model has uncovered built-in censorship mechanisms that limit discussions on topics considered politically sensitive by the Chinese government, such as the 1989 Tiananmen Square massacre and the ongoing treatment of Uyghurs. This raises concerns about the suppression of free speech and how political parties manipulate state-influenced narratives through AI platforms.
Data privacy and security
DeepSeek states that user data, including text inputs, uploaded files, and chat histories, are stored on servers within China and shared with third-party users like Baidu and ByteDance. This raises concerns about user privacy, data security and the possibility of State surveillance.
Given the U.S.’s concerns about ByteDance, it is no surprise that the U.S. National Security Council is investigating DeepSeek’s data collection and usage policies.
Intellectual Property and Fair Use
Developing AI models usually involves training on vast datasets that include copyrighted materials. This practice raises questions about the legality and ethics of using copyrighted material without the creator's explicit permission.
While DeepSeek represents a significant advancement in AI technology, it continuously underscores AI's ethical considerations, such as censorship, data privacy, intellectual property rights, national security, and the challenges associated with open-source technology. Addressing these issues is crucial for ensuring that AI is utilised constructively and responsibly globally.
While DeepSeek represents a significant advancement in AI technology, it continously underscores AI's ethical considerations, such as censorship, data privacy, intellectual properety rights, national security, and the challenges associated with open-source technology.
DeepSeek's arrival presents an opportunity for AI technology to shift towards a more open, decentralised landscape. If the opportunities and risks presented by DeepSeek are well-managed, it could lead to a more innovative and competitive AI landscape that benefits both the industry and the user. However, mismanaging them will increase geopolitical tensions between China and the U.S. and potentially accelerate the ethical and security dilemmas inherent within AI. One thing is certain: the impact of DeepSeek will be a catalyst for ongoing dialogue and development within the AI ecosystem.
[1] Booth, R., & Milmo, D. (2025, February 7). Into the deep. The Guardian Weekly.
[2] Malik. K. (2025, February 7). Machine Codes DeepSeek has ripped away the veil of mystique around AI. The Guardian Weekly.
