Elice’s first two trading sessions in Seoul offered a compressed version of the argument now gripping the AI market: enormous infrastructure demand on one side, mounting concern over who will ultimately pay for it.

On Tuesday, the South Korean neocloud provider raised 201 billion won ($151 million) in an IPO priced at the top of its range, then saw its shares plunge 32% in their market debut. The sell-off came as investors reassessed companies building specialized cloud capacity for AI workloads, particularly those whose expansion depends on outside financing.¹

The stock rebounded roughly 30% on Wednesday, but the recovery did not erase the warning embedded in the debut. Dan Ives of Yorkville Ives called it a “gut check moment in the tech trade,” arguing that investors are increasingly sorting potential AI winners from “weak links” in the sector’s food chain.¹

Other analysts see a more structural shift. Billy Leung of Global X ETFs said the market has moved beyond applauding the sheer scale of capital-expenditure announcements and is now asking “who can actually earn an adequate return on that capital.” In that view, companies generating cash during the build-out are pulling away from those that need fresh capital simply to continue it.¹

John Blank of Zacks Investment Research framed the pressure as broader than speculation, pointing to credit stress and warning that AI and semiconductor investment may be especially overbuilt in South Korea. Leung, however, said the regional pullback appears driven more by worsening sentiment and lower valuations than by a sudden collapse in AI demand.¹

The emerging consensus is not that the AI boom is over. It is that lenders and shareholders are demanding proof: solid tenant contracts, credible pricing and a path to profitability before financing another costly data-centre expansion.¹