((Automated translation by Reuters, please see disclaimer https://bit.ly/rtrsauto))
(Updated with company statement in paragraph 4)
Nvidia NVDA.O's new Blackwell AI chips, which have already suffered delays, have encountered issues with overheating servers that accompany them, leading some customers to worry that they won't have enough time to implement. new data centers, The Information newspaper reported on Sunday.
Blackwell graphics processors overheat when connected together in server racks designed to hold up to 72 chips, the report said, citing sources familiar with the problem.
The chipmaker has asked its suppliers to change rack designs multiple times to address overheating issues, according to Nvidia employees who worked on the issue, as well as customers and suppliers with knowledge of the problem, a indicated the report without naming the suppliers.
“Nvidia works with leading cloud service providers as an integral part of our engineering team and process. Engineering iterations are normal and expected,” a company spokesperson told Reuters.
In March, Nvidia unveiled the Blackwell chips and previously said they would ship in the second quarter before encountering delays, potentially affecting customers such as Meta Platforms META.O , Alphabet's GOOGL.O Google and Microsoft MSFT.O .
Nvidia's Blackwell chip takes two squares of silicon the size of the company's previous offering and bonds them into a single component that's 30 times faster for tasks like chatbots providing answers.