DeepSeek’s AI takeover: Is US tech falling behind?

Getting your Trinity Audio player ready...

DeepSeek, the China-based artificial intelligence (AI) company, has been a disrupting force in the AI market. The company was founded in 2023 but flew under the radar until January 10, 2025, when it launched the DeepSeek chatbot powered by its DeepSeek R1 model. DeepSeek became OpenAI’s top competitor within two weeks and the most downloaded free app on Apple’s App Store.

How DeepSeek skyrocketed to the top

Two key factors fueled DeepSeek’s rapid rise. First—and most importantly—was the media frenzy around DeepSeek R1. The company claims it built a model just as capable as OpenAI’s GPT-o1 but at a fraction of the cost: just $5.58 million, using 2,000 Nvidia (NASDAQ: NVDA) H800 GPUs, which is many multiples below the billions of dollars that AI tech giants in the U.S. are spending. That got the media and the tech industry talking about DeepSeek, and they haven’t stopped since.

Second, developers are moving toward DeepSeek because it’s more affordable than OpenAI. Running AI models on DeepSeek is significantly cheaper, which makes it attractive to startups and independent developers.

DeepSeek’s rise made markets spiral

For years, the U.S. led the global AI race. Companies like OpenAI, Google DeepMind (NASDAQ: GOOGL), and Anthropic have been dominant players in the space. But DeepSeek’s sudden rise has people asking: if a new player from China can do this for under $6 million, why are U.S. companies spending billions? Are U.S. tech companies overvalued? Should we begin to allocate investment dollars to overseas companies like DeepSeek?

But the questions and conversation go beyond the media frenzy taking place. Investors have taken note of this as well. On January 27, when DeepSeek’s popularity seemed to have peaked, U.S. stock markets took a hit. The S&P 500 (NASDAQ: SPX) closed down about 1.5%, Nasdaq closed the day down around 3%, and Nvidia—despite being a key supplier of GPUs for DeepSeek—plummeted nearly 18% amid fears that the implications of an allegedly cheaper-to-build, more efficient model would upend companies in the United States.

DeepSeek’s threat to US national security

Beyond DeepSeek’s financial implications, it is setting off national security alarms in the United States. The American government has long distrusted Chinese technology. That’s one of the reasons it restricted AI chip exports to China—to curb its advancements in AI.

But now, with DeepSeek going live and becoming so popular, even among U.S. residents, there are growing fears that DeepSeek is collecting data from U.S. users, including intellectual property, keystroke patterns, and other metadata, and that this data could be weaponized, giving China an edge in AI, cultural, and economic warfare.

World governments and government organizations are already acting on these fears. Italy has pulled DeepSeek from the Apple App Store and Google Play Store, and the U.S. Navy has outright banned its use, citing security risks.

U.S. NAVY BANS USE OF DEEPSEEK DUE TO ‘SECURITY AND
ETHICAL CONCERNS’ – CNBC

— *Walter Bloomberg (@DeItaone) January 28, 2025

Did the AI market overreact?

The market reaction to DeepSeek’s launch is overblown. Nvidia’s stock drop, for example, doesn’t make sense; DeepSeek uses Nvidia chips, and its success will only drive more demand for them.

However, U.S. AI firms can’t be blind to this new competitor in the market. DeepSeek has proven that AI models don’t have to cost billions to build. This forces U.S. tech CEOs to rethink their strategies; even President Donald Trump weighed in, calling DeepSeek a “wake-up call” for the American AI industry.

But it’s important to note that just because DeepSeek has been successful doesn’t mean it’s groundbreaking. Unlike OpenAI and others that had to build AI from scratch, DeepSeek entered the scene when much of the foundational AI research was already done, which makes it easier—and cheaper—to develop an AI model.

Did DeepSeek copy OpenAI?

One of the latest unanswered questions is whether DeepSeek built its model from scratch or if it trained on OpenAI’s proprietary data.

Microsoft (NASDAQ: MSFT), a major OpenAI investor, is now investigating whether DeepSeek illegally used OpenAI research. If that turns out to be true, DeepSeek could face legal action and regulatory crackdowns.

Obviously

— Elon Musk (@elonmusk) January 27, 2025

Some individuals, including Elon Musk, also argue that DeepSeek’s reported development costs are too low to be real. They suspect China found ways to get around U.S. chip restrictions, securing more powerful resources than it admits and obfuscating the actual development cost so that it doesn’t look like the country violated the restrictions.

DeepSeek’s future

The security concerns around DeepSeek are real. But for everyday users, it’s probably just another AI chatbot. Developers will stick with it because it’s cheap, but for the general public, DeepSeek is just a tech product that got hyped up and induced market panic—I think the hype will die out pretty quickly unless DeepSeek proves itself. If it isn’t significantly better than the competitors in the market—which it isn’t at the moment—then I doubt the average user will be switching over from their current AI provider to this new competitor.

The real test is going to be whether DeepSeek actually innovates. If it beats OpenAI to the next big breakthrough, it will gain credibility as a serious contender in the global AI race. But if it keeps rolling out similar models after OpenAI does, the “copycat” accusations will stick.

In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.

Watch Alex Ball on the future of tech: AI development and entrepreneurship