16
DeepSeek and the Stock Market
A Chinese AI company, DeepSeek, is making waves in the US and on Wall Street. Their new technology is so impressive that some experts think it’s as good as OpenAI’s ChatGPT.
What’s really surprising investors is how cheaply DeepSeek developed their AI app. Experts estimate it only cost them about $6 million. That’s a tiny fraction of what big American companies like OpenAI and Google are spending. These US companies are planning to invest around a trillion dollars in AI over the next few years, according to financial analysts Goldman Sachs.
DeepSeek, a Chinese tech company, recently launched something new that caused the stock prices of companies like Nvidia and ASML (both big chip makers) to drop on Monday. These companies make the specialized chips needed for AI. DeepSeek’s technology has people wondering if there might be less demand for Nvidia’s chips in the future. It also makes some experts question whether investors have been too excited about tech stocks related to AI, like Meta and Microsoft.
According to Giuseppe Sette from Reflexivity, DeepSeek’s new product is impressive because it achieves more with fewer resources. He believes this shows that the AI world will continue to see unexpected developments.
This news about DeepSeek comes shortly after a former US president announced a new $500 billion project called Stargate with OpenAI (the maker of ChatGPT), Softbank, and Oracle. He claimed this project would guarantee the future of technology in the US.
What is DeepSeek?
DeepSeek, a new Chinese tech company, was started in July 2023 by Liang Wenfeng. Liang, a graduate of a top Chinese university, used money from his own investment fund to get the company going, according to MIT Technology Review. He’s a very wealthy man with around $8 billion, according to a financial analyst named Ives.
Liang used to work in finance, using AI to help with investments. Before certain computer chips were banned from being sold to China, he bought a lot of Nvidia A100 chips. These chips are now the foundation of DeepSeek, the MIT report said.
Another expert, Ben Reitzes from Melius, told investors that DeepSeek has made real progress in AI. He said their AI tool learns better and uses computer memory more efficiently. However, Reitzes also questioned whether they really needed so many of those chips.
Privacy issues with DeepSeek?
DeepSeek is facing significant privacy concerns due to its data collection practices and its obligations under Chinese law. Here’s a breakdown of the key issues:
Extensive Data Collection:
1. DeepSeek’s privacy policy states that it collects a wide range of personal data, including:
2. Profile information (name, email, phone number, etc.)
3. User input (everything you type or upload, including chat history)
4. Device and network data (IP address, device model, etc.)
5. Usage data (features you use, actions you take)
6. Cookies and trackers
Third-party data (information from linked accounts
What does DeepSeek mean for Nvidia and other tech companies?
Wall Street is trying to assess the long-term impact of a low-cost AI tool from China that rivals ChatGPT and other so-called generative AI apps. It also raises questions about whether Silicon Valley is overspending on tech advancements in the AI sector, noted Angelo Zino, senior equity analyst at CFRA Research, in an email.
“The fact that this technology is supposed to take less energy and is more cost-effective than U.S.-based models has U.S. technology investors very concerned,” Jay Woods, chief global strategist at Freedom Capital Markets, said.
It’s also unclear what type of pushback or reaction could come from the White House, given that Mr. Trump has raised the possibility of placing new tariffs on Chinese imports, although he also gave the Chinese-owned TikTok a reprieve by ordering the Justice Department not to enforce a looming ban.
Some Wall Street analysts think Monday’s stock selloff is an overreaction, noting that the enormous demand for AI will continue lifting key players in the sector.
“It’s one thing to train a [large language] model for less money, but accommodating the huge demand for the consumption of all this AI technology is still going to require massive amounts of infrastructure,” Adam Crisafulli of VitalKnowledge said in a report.
Analysts at Bernstein Research also noted that while DeepSeek’s technology looks promising, it likely isn’t revolutionary, suggesting that Monday’s investment rout is overblown.
What is Nvidia saying about DeepSeek?
Nvidia told CBS News that they’re impressed with DeepSeek. They called it “an excellent AI advancement” and a great example of “test-time scaling.” In an email, Nvidia explained that DeepSeek shows how new AI models can be built using this technique, using common models and readily available computing power that complies with export rules.
However, Nvidia also pointed out that using AI models to make decisions (called “AI inference”) needs a lot of Nvidia graphics cards and fast networking. They added that there are now three ways to scale AI: pre-training and post-training (which are ongoing), and this new “test-time scaling.”
