Deepseek AI News Live Updates: Tech stock sell-off sparked by DeepSeek deepens on Day 2

AI chip leader Nvidia closed at 8.9% on Tuesday after falling by 17 per cent and losing $593 billion in market value a day prior, according to a report by Reuters.

Oracle finished up 3.6% after falling about 13.8% on Monday. And Marvell Technology shares rose 3.5% after falling 19% the previous day. Shares of another chip heavyweight, Broadcom, gained 2.6% on Tuesday after dropping 17.4% on Monday, the report said.

“DeepSeek’s breakthrough in AI model development, leveraging widely available resources, represents a paradigm shift in how artificial intelligence can be created and deployed. This approach directly challenges the narrative of U.S. technological exceptionalism, particularly the dominance of the ‘Magnificent Seven’ tech giants,” said Devarsh Vakil – Head of Prime Research – HDFC Securities.

Such disruption raises concerns for the NASDAQ index, where these companies’ outsized market capitalizations, worth trillions of dollars, heavily influence market performance, he said.

Global technology shares sank on Tuesday, as a market rout sparked by the emergence of low-cost AI models by DeepSeek entered its second day, according to a report by Reuters.

Some of Japan's biggest tech companies came under pressure for a second day such as chip-testing equipment maker Advantest (down 10%) and tech start-up investor SoftBank Group (down 5%), the report said, adding that a number of Big Tech companies, including Apple and Microsoft, are expected to report earnings this week.

David Sacks, US President Donald Trump's AI and crypto adviser, said DeepSeek's success justified the White House's decision to roll back former US President Joe Biden's AI policies.

"DeepSeek R1 shows that the AI race will be very competitive and that President Trump was right to rescind the Biden EO, which hamstrung American AI companies without asking whether China would do the same. (Obviously not.) I’m confident in the U.S. but we can’t be complacent," Sacks said in a post on X.

How did a tech startup backed by a Chinese hedge fund manage to develop an open-source AI model that rivals our own? That’s what Meta CEO Mark Zuckerberg has set out to determine by assembling four teams of engineers, according to a report by The Information.

The report detailed Meta’s efforts to catch up to DeepSeek whose open-source technology has called into question the massive investments made by American companies like Meta on AI chips.

While Meta may be in high-alert mode behind doors, its chief AI scientist insists that DeepSeek’s breakthrough is ultimately good news for the social media giant.

“To people who see the performance of DeepSeek and think: ‘China is surpassing the US in AI.’ You are reading this wrong. The correct reading is: ‘Open source models are surpassing proprietary ones,’” Yann LeCun said in a LinkedIn post.

“DeepSeek has profited from open research and open source (e.g., PyTorch and Llama from Meta). They came up with new ideas and built them on top of other people’s work. Because their work is published and open source, everyone can profit from it. That is the power of open research and open source,” he added.

A large-scale cyber attack targeting DeepSeek has caused it to temporarily limit user registrations. Confirming the cybersecurity incident, the Chinese AI startup said it is assessing the extent of the cyber attack and taking precautionary steps to mitigate any further damage.

Meanwhile, social media users questioned the security of user data maintained by DeepSeek and the integrity of its AI chatbot service.

After Nvidia’s shares tumbled 17 per cent, cutting nearly $600 billion from its market cap, the chipmaker released a statement on Monday.

“DeepSeek is an excellent AI advancement and a perfect example of Test Time Scaling. [...] DeepSeek’s work illustrates how new models can be created using that technique, leveraging widely-available models and compute that is fully export control compliant," Nvidia said in a statement.

“Inference requires significant numbers of NVIDIA GPUs and high-performance networking. [...] We now have three scaling laws: pre-training and post-training, which continue, and new test-time scaling,” an Nvidia spokesperson was quoted as saying by CNBC on Monday.

Since 2022, the US government has imposed restrictions on the export of AI hardware to China. These export controls have been openly criticised by Nvidia in the past. Despite the lack of access to the latest AI hardware, DeepSeek developed AI models rivaling o1 and Gemini using far lesser Nvidia chips than its US counterparts.