innovation, resources
DeepSeek Reopens API Access Amidst Growing AI Competition in China
DeepSeek has reopened API access after a three-week suspension, resuming services amid rising AI competition in China. Alibaba and ByteDance are advancing AI models, while U.S. chip restrictions push China towards self-sufficiency. Dr. Serena Huang, a Chief Data Officer and AI consultant for Fortune 100 companies, highlight ethical concerns, AI biases, and the need for transparent governance to address AI-driven misinformation and censorship.
DeepSeek, a Chinese artificial intelligence company that has gained significant traction since its launch, has resumed access to its core programming interface after nearly a three-week suspension. The company, which emerged as a key player in AI development last month, had halted credit top-ups in early February due to capacity shortages. Now, with services reinstated, DeepSeek has reassured users that while credits can be purchased again, server resources will remain constrained during peak hours.
This move marks a crucial step in the adoption of DeepSeek’s AI model, which has been widely used by Chinese businesses across various sectors. The platform has drawn comparisons to OpenAI’s ChatGPT, with many noting its competitive capabilities developed at a fraction of the cost of leading AI products. The reopening of API access comes amid an intensifying AI race in China, with companies striving to solidify their positions in the global market.
As China accelerates its AI advancements, applications of AI are expanding beyond business and research—products like AI Girlfriend are gaining popularity, leveraging sophisticated language models to create interactive and emotionally responsive virtual companions.
Competitive landscape and Alibaba’s strategic investments
The resumption of DeepSeek’s services coincides with Alibaba Group Holding Ltd.’s unveiling of its latest AI model, QwQ-Max. The model’s preview underscores the increasing competition in China’s AI sector, as companies aim to accelerate their advancements in the field.
In addition, ByteDance Ltd., the parent company of TikTok, is currently testing a deep reasoning AI model with a limited group of users, further adding to the growing list of competitors in the AI space.
Alibaba has also made a substantial financial commitment to AI and cloud computing. The company recently announced a $53 billion investment over three years to strengthen its infrastructure, a major shift for the e-commerce giant. On Tuesday, it declared plans to open-source QwQ-Max, a decision that could shape the industry’s direction and influence global AI research.
China’s role in the global AI Arms Race
The advancements made by DeepSeek and other Chinese AI firms highlight the country’s growing influence in the global AI landscape. Dr. Serena Huang, a Chief Data Officer and AI consultant for Fortune 100 companies, has weighed in on the broader implications of these developments.
“The U.S. chip export restrictions may have some short-term impact on China’s AI progress, but they are unlikely to be effective in the long run,” Dr. Huang noted. “China has been aggressively investing in self-sufficiency, and the development of alternative chip architectures, such as those used in DeepSeek R1, suggests that China is making strides in reducing its reliance on U.S. chip exports.”
She further explained that rather than hindering progress, these restrictions might accelerate China’s domestic innovation, pushing companies to develop independent solutions.
“This could ultimately lead to China emerging as a leader in AI chip design and manufacturing,” she added.
Dr. Huang also emphasised the importance of global collaboration and ethical considerations in AI development. “Western companies and governments should acknowledge China’s rapid progress in AI and recognise the potential for China to shape the future of AI. To remain competitive, they should invest in AI research and development, attract and retain top talent, and prioritise ethical considerations in the development and deployment of AI technologies.”
She further stressed the necessity of establishing international standards and governance protocols to ensure ethical AI usage. “By doing so, they can ensure that the development and use of AI technologies are aligned with democratic values and human rights principles,” she stated.
DeepSeek’s approach to AI and content moderation challenges
Despite its rapid growth, DeepSeek has faced scrutiny regarding the implications of its AI capabilities. According to Dr. Huang, AI-generated content has the potential to reinforce political narratives and limit the accessibility of diverse perspectives.
“DeepSeek AI is being used not just as a tool for generating data and insights but can become a means of reinforcing political narratives, limiting what people can access and learn,” she explained. “This form of censorship poses a clear challenge: how do we ensure that the content generated by AI is unbiased, fair, and representative of a diverse range of viewpoints?”
She cautioned against relying solely on free AI tools, highlighting concerns about transparency and control over AI-generated content. “It’s more important than ever to not fall for ‘free’ tools and consider who’s investing in the company and what you are giving up in exchange for this ‘free’ tool. Because really, nothing is free.”
Dr. Huang underlined the importance of critical thinking in navigating the evolving AI landscape. “In an era of potential AI censorship, critical thinking becomes even more important to ensure we can detect misinformation and manipulation. Companies need to recognise that AI should not be seen as a replacement for human judgment and empathy but rather as a complement to human skills.”
DeepSeek’s open-source initiative and the future of AI development
In response to growing interest in its technology, DeepSeek has announced plans to release key code and data to the public. This step, considered unusual among major AI developers, sets DeepSeek apart from its competitors, including OpenAI, by making portions of its core technology openly accessible.
On Wednesday, DeepSeek stated on X that it was making available the “DeepGEMM” library, a component used in the training and inference processes of its V3 and R1 AI models. This move signals the company’s intention to contribute to the broader AI research community while intensifying competition in the AI arms race between China and the U.S.
As China’s AI sector continues to expand, industry observers anticipate further advancements from DeepSeek and its competitors. With increased investment, regulatory considerations, and global collaboration shaping the future of AI, the technology’s development and its implications on society remain critical topics for discussion.
About Dr. Serena Huang
Dr. Serena Huang is a leading AI consultant and Chief Data Officer, specialising in the application of data and AI to organisational strategy, wellbeing, and diversity. With over a decade of experience in analytics leadership at global firms such as GE, Kraft Heinz, and PayPal, she has played a key role in shaping AI-driven decision-making processes.
She is an international keynote speaker, a LinkedIn Learning instructor, and the founder of Data with Serena, an advisory firm dedicated to helping organisations harness AI for ethical and effective business solutions. Dr. Huang is also the author of The Inclusion Equation – Leveraging Data & AI for Organisational Diversity and Wellbeing.
