U.S. markets tumbled Monday morning following concerns over a breakthrough from DeepSeek. The Chinese AI startup claimed to have trained an AI model on par with advanced U.S. technology at a fraction of the cost. DeepSeek reported over $5 million in training expenses. OpenAI’s GPT-4 model costs more than $100 million to train, and Anthropic CEO Dario Amodei said the cost to train some models is approaching $1 billion. DeepSeek’s claims about creating a low-cost AI model face skepticism from analysts, who question the feasibility of achieving such results without advanced GPUs and raise doubts about the accuracy of the company's assertions.
Category
🗞
NewsTranscript
00:00It's Benzinga, bringing Wall Street to Main Street.
00:02U.S. markets tumbled Monday morning following concerns over a breakthrough from DeepSeek.
00:07The Chinese AI startup claimed to have trained an AI model on par with advanced U.S. technology
00:14at a fraction of the cost. DeepSeek reported over $5 million in training expenses.
00:19OpenAI's GPT-4 model costs more than $100 million to train. Anthropic CEO Dario Amati
00:26said that the cost to train some models is approaching $1 billion.
00:30DeepSeek claims about creating a low-cost AI model face skepticism from analysts,
00:35who question the feasibility of achieving such results without advanced GPUs,
00:41and raise doubts about the accuracy of the company's assertions.