DeepSeek Eyes $45B Valuation in State-Backed First Funding Round

China's National Integrated Circuit Industry Investment Fund — commonly known as the 'Big Fund' — is in active negotiations to lead DeepSeek's first-ever external funding round, potentially valuing the AI startup at approximately $45 billion. The valuation has roughly doubled from the $20 billion figure discussed just weeks ago, signaling extraordinary investor confidence in the company that rattled Silicon Valley earlier this year.

The deal, first reported by the Financial Times on May 6, would mark a landmark moment for both DeepSeek and the Chinese government's strategic investment apparatus. If finalized, it represents the Big Fund's first public investment in a domestic large language model company — a significant pivot from its traditional focus on semiconductor hardware.

Key Takeaways

• Valuation surge: DeepSeek's estimated worth jumped from ~$20 billion to ~$45 billion in just a few weeks of negotiations
• State backing: China's Big Fund, a government-backed semiconductor investment vehicle, is negotiating to lead the round
• Tech giant involvement: Tencent is among other potential investors in discussions
• Founder commitment: Liang Wenfeng, who controls ~89.5% of DeepSeek, may personally invest in this round
• Strategic first: This would be the Big Fund's first-ever public investment in an LLM company
• Pre-revenue bet: Investors are backing DeepSeek despite its limited commercialization efforts to date

Valuation Doubles in Weeks as Investor Frenzy Intensifies

The speed at which DeepSeek's valuation has climbed is remarkable even by AI industry standards. During early-stage negotiations just weeks prior, external observers pegged the company's worth at around $20 billion — already a substantial figure for a company with minimal commercial revenue.

Now, that number has effectively doubled to $45 billion (approximately 307.8 billion Chinese yuan at current exchange rates). The leap underscores how aggressively investors are competing for a stake in what many consider China's most promising AI lab.

For context, this valuation places DeepSeek in rarefied air alongside some of the world's most valuable AI startups. Anthropic, the maker of Claude, was valued at $61.5 billion in its most recent funding round. OpenAI sits at a reported $300 billion. While DeepSeek trails both, its ascent from relative obscurity to a $45 billion valuation — without a single prior external funding round — is virtually unprecedented.

Why the Big Fund's Involvement Changes Everything

The National Integrated Circuit Industry Investment Fund was established by the Chinese government to bolster the country's domestic semiconductor capabilities. In 2024 alone, it raised approximately $47 billion from the Ministry of Finance, local governments, and state-owned banks.

Historically, the Big Fund has focused exclusively on hardware — pouring capital into chip manufacturers like SMIC (Semiconductor Manufacturing International Corporation) and memory chip maker Yangtze Memory Technologies. These investments target the physical infrastructure needed to reduce China's dependence on foreign semiconductor suppliers, particularly amid escalating U.S. export controls.

A move into large language model investment would represent a significant strategic expansion. It signals that Beijing increasingly views frontier AI models — not just the chips that power them — as critical national infrastructure.

This shift carries profound implications:

• It suggests the Chinese government considers DeepSeek a national strategic asset
• It blurs the line between hardware and software in China's tech self-sufficiency strategy
• It could accelerate government support for AI model development across the ecosystem
• It may trigger similar state-backed investments in other Chinese AI labs

Tencent and Liang Wenfeng Join the Investor Mix

Beyond the Big Fund, other notable names have emerged in the funding discussions. Tencent, one of China's largest technology conglomerates and the parent company of WeChat, is reportedly among the potential co-investors. Tencent has been steadily building its AI portfolio, and a stake in DeepSeek would give it access to one of the most talked-about model developers in the world.

Perhaps more intriguing is the involvement of Liang Wenfeng, DeepSeek's founder and controlling shareholder. Sources indicate that Liang plans to personally invest in this round. He currently holds approximately 89.5% of the company, maintaining near-total control over its direction.

Liang's decision to invest his own capital alongside institutional investors sends a strong signal of confidence. It also raises questions about how the ownership structure might evolve — and whether DeepSeek will maintain its famously independent, research-first culture as external shareholders enter the picture.

The final roster of investors has not been confirmed. Negotiations remain fluid, and the composition of the round could still shift before a deal is formally announced.

DeepSeek's Rise: From Hedge Fund Side Project to Global AI Contender

DeepSeek's trajectory has been one of the most surprising stories in AI over the past year. Founded by Liang Wenfeng — who also runs the quantitative hedge fund High-Flyer — the company burst onto the global stage in January 2025 with its DeepSeek-R1 reasoning model.

What made DeepSeek's models so disruptive was not just their performance but their efficiency. The company claimed to have trained competitive large language models at a fraction of the cost typically associated with frontier AI development. DeepSeek-R1 demonstrated reasoning capabilities that rivaled offerings from OpenAI and Google, despite being developed with reportedly fewer high-end GPUs and at dramatically lower training costs.

The implications sent shockwaves through the AI industry:

• Nvidia's stock briefly dropped as investors questioned whether the GPU arms race was as critical as assumed
• Silicon Valley was forced to reckon with a credible Chinese competitor achieving parity at lower cost
• Open-source advocates celebrated DeepSeek's decision to release model weights publicly
• U.S. policymakers intensified debates about export controls and their effectiveness

Despite all the attention, DeepSeek has remained largely focused on research rather than aggressive commercialization. The Financial Times noted that investors are backing the company's potential rather than current revenue — a dynamic familiar in the Western AI startup ecosystem but notable given the scale of this particular bet.

How This Compares to Western AI Funding

The global AI funding landscape has been defined by staggering numbers. OpenAI has raised over $30 billion in total funding. Anthropic has secured more than $15 billion from investors including Amazon and Google. xAI, Elon Musk's AI venture, closed a $6 billion round in late 2024.

DeepSeek's potential $45 billion valuation in its first external round places it immediately among the top tier of global AI companies by valuation. No Western AI startup achieved a comparable valuation at the debut funding stage.

This comparison highlights several important dynamics. First, it reflects the intense geopolitical competition driving AI investment on both sides of the Pacific. Second, it demonstrates that state-backed capital can move as aggressively — if not more so — than Silicon Valley venture capital. Third, it raises questions about valuation sustainability in an industry where revenue models remain uncertain for many players.

What This Means for the Global AI Race

For Western AI companies and policymakers, DeepSeek's massive state-backed funding round is a wake-up call. The investment signals that China is not merely competing in the AI race — it is mobilizing sovereign capital to ensure its champions have the resources to compete at the frontier.

For developers and businesses worldwide, the implications are nuanced. A well-funded DeepSeek could accelerate open-source model development, potentially driving down costs and increasing accessibility for the global developer community. However, it also raises questions about data governance, model safety, and the geopolitical complexities of relying on Chinese-developed AI infrastructure.

For investors, the deal represents both opportunity and risk. The AI valuation environment is increasingly frothy, with limited revenue to justify multi-billion-dollar price tags. DeepSeek's minimal commercialization makes this a pure bet on future potential — the kind of wager that can generate enormous returns or significant losses.

Looking Ahead: What Comes Next for DeepSeek

Several critical questions remain as negotiations continue. The final valuation, investor lineup, and deal structure are all still in flux. Key developments to watch include:

• Deal closure timeline: No formal announcement date has been set, and terms could shift
• Commercialization strategy: How DeepSeek plans to generate revenue will become a central question for new investors
• Regulatory dynamics: U.S. and European responses to state-backed AI investment in China may intensify
• Model releases: Whether continued open-source releases align with investor expectations for monetization
• Talent competition: A well-funded DeepSeek could attract top AI researchers from both Chinese and Western institutions

If the $45 billion valuation holds and the Big Fund formally leads the round, it will mark a defining moment in the global AI landscape. The message is clear: the competition for AI supremacy is no longer just a battle between startups and tech giants — it is increasingly a contest between nations and their sovereign investment strategies.

DeepSeek's first funding round is not just a financing event. It is a geopolitical signal.