Sandisk forecasts profit surge, secures supply deal as AI fuels storage demand

By Stephen Nellis

SAN FRANCISCO, Jan 29 (Reuters) – Sandisk on Thursday predicted profits and sales well above Wall Street estimates and extended a major supply agreement, riding a surge in artificial-intelligence-driven demand for data storage.

The data storage firm forecast fiscal third-quarter revenue with a midpoint of $4.6 billion and adjusted profit with a midpoint of $14 per share. Both were above estimates of $2.77 billion and $4.37 apiece, respectively, according to LSEG-compiled data.

Silicon Valley-based Sandisk supplies flash storage memory, the basis for solid-state drives that hold massive amounts of data inside AI data centers. While most of a burgeoning global shortage of memory chips has focused on DRAM, the faster type of memory chip that sits closer to a computer’s processor, AI is also increasing demand for flash storage, Sandisk CEO David Goeckeler told Reuters.

Large AI firms are building data centers for what is known as “inference,” when AI models answer questions from users, a  process that requires feeding stored data into computing chips. Sandisk’s forecast reflects the fact that those firms are willing to spend money amid tight flash storage supplies to keep their plans on track.

“Customers prefer supply over price,” Goeckeler said.

Sandisk reported sales and adjusted profits of $3.3 billion and $6.20 per share for the just-ended fiscal second quarter, above estimates of $2.64 billion and $3.33 per share, respectively, according to LSEG-compiled data.

Sandisk secures its flash chip supply through a joint venture with Kioxia Corp in Japan.

“We have incredible capacity in Japan that we’ve been investing in, and we continue to invest in. Now we’re signed up together for another nine years,” Goeckeler said.

“So that’s our capacity plan going forward.”

The companies said that they have extended their supply agreement through the end of 2034, from its previous expiration at the end of 2029.

(Reporting by Stephen Nellis in San Francisco; Editing by Sherry Jacob-Phillips)