By Selena Li and Sumeet Chatterjee
HONG KONG, Nov 17 (Reuters) - Citigroup is more than halfway through its plan to hire roughly 100 bankers for its global corporate banking unit and expects strategic AI-related
capital expenditure to remain on track despite market concerns about spending, a senior executive said.
Among the hires already made are more than 40 bankers for its cross-border corporate banking services, Kaleem Rizvi, Citi's global co-head of corporate banking, told Reuters, as supply chains shift, new trade channels emerge and demand surges for financing to support the global AI build-out.
"We'll invest more and look to scale it up even further," Rizvi said. "Investment in talent, in bankers, is key."
A bruising selloff in AI-related stocks, prompted by the resignation of Nvidia's CEO, Jensen Huang, has put a new focus on the sector's capital expenditure plans and how much AI exposure investors and financial firms have.
A key moment for markets on Tuesday will be Nvidia's earnings, with investors watching for signs of whether its debt-fuelled AI spending will continue into 2027 and how Huang's departure could affect its influence across the AI ecosystem.
Rizvi said while AI-related demand would continue to grow, there would be a "sharper focus" on return on equity and strategic capital expenditure, which, he said, was good for the markets.
"It gives people a pause and a breather to focus on executing what they're already spending (on)," he said. "I see it as a pause, a bit of a slowdown."
Major Asian suppliers to the AI boom, such as South Korea and Taiwan, will benefit from the rising capex "for the foreseeable future", according to Rizvi.
Credit stress "could manifest at some point" in the broader financial sector, but it was too early for the current selloff in the AI sector to affect banks' balance sheets in a material way, Rizvi said.
Rizvi said Asia should be able to capture a significant share of any redirected investment in US data centres as public pushback grows following the US midterm elections.
However, investments in Asia face tighter US export rules aimed at preventing Chinese firms from accessing advanced AI chips, including via third-country data centres.
It was too early to say how the situation would play out, Rizvi said, adding that Chinese hyperscalers are expected to lean on domestic chip producers to secure their computing power.
(Reporting by Selena Li and Sumeet Chatterjee; Editing by John Mair)












