DeepSeek, an AI startup based in Hangzhou, China behind the R1 model, has restricted signups to users with mainland China telephone numbers.
The company attributed this decision to “large-scale malicious attacks” on its services, following an unprecedented surge in demand for its AI assistant.
The sudden popularity of DeepSeek’s AI assistant propelled the app to the top of Apple’s iPhone download charts and among the most downloaded on Google’s Play Store.
However, this surge overwhelmed the startup’s systems, leading to its longest reported outage.
DeepSeek stated that signups are now limited to users with mainland China mobile number.
“Currently, only registration with a mainland China mobile phone number is supported,” the company stated.
The company has not clarified whether this limitation is temporary or how long it will last.
AI dominance
In just 36 hours, the company’s AI assistant, powered by its R1 model, surged to the top of app download charts, sparking a wave of interest from tech heavyweights and investors alike.
- The company’s AI model quickly climbed the ranks on a UC Berkeley-affiliated AI leaderboard, signaling its growing potential.
- Within a short period, the startup’s AI assistant rocketed to the top of Apple’s iPhone download charts and ranked among the most downloaded apps on Google’s Play Store.
DeepSeek’s AI model sets itself apart from competitors by offering transparency in its responses. Unlike other models, DeepSeek shows its reasoning and work as it addresses user queries, a feature that has received praise on both Apple and Google app stores.
Impact on global markets
The rapid rise of DeepSeek’s AI assistant had significant ramifications on global markets. The company’s claims of offering a more cost-efficient AI solution triggered a $1 trillion rout in US and European technology stocks.
Investors began to question the sustainability of AI spending, particularly in light of DeepSeek’s ability to develop a high-performing AI model at a fraction of the cost of its rivals.
“While it remains to be seen if DeepSeek will prove to be a viable, cheaper alternative in the long term, initial worries are centered on whether US tech giants’ pricing power is being threatened and if their massive AI spending needs re-evaluation,” Jun Rong Yeap of IG Asia stated.
Challenging AI assumptions
DeepSeek’s success challenges the prevailing assumption that advanced AI requires vast amounts of computing power and energy.
- While the startup has not fully disclosed its development and energy costs, it has suggested that its model was developed at a fraction of the cost of OpenAI’s latest models.
- This efficiency has raised questions about the effectiveness of US trade sanctions on advanced technology, particularly in light of DeepSeek’s emergence despite restrictions on Nvidia chips.
However, DeepSeek’s self-censorship on politically sensitive topics in China, such as Tiananmen Square, President Xi Jinping, and Taiwan, has raised concerns among international users. This censorship may present challenges for DeepSeek as it looks to expand beyond China.
