Google DeepMind CEO Demis Hassabis Says Memory Chip Shortages Are Creating A ‘Choke Point’ For AI

As we turned from 2025 to 2026, the world of AI has gotten even more sophisticated and integrated with our lifestyle; only recently Google dropped its Gemini 3.1 Pro, its latest in-line AI suite capable of better core reasoning and multitasking skills. Anthropic has its Opus, and others are competing in this race as well, but Google DeepMind's CEO Demis Hassabis thinks this is all going to come to a standstill, and memory could be the reason behind it.

Memory shortages are becoming a reality right now as big tech giants capture the major chunk of the produce to keep their hefty data centers working. Hoards of GPUs and chips are being used to run these AI models, which for now are being mostly utilized for generating pop culture art-inspired personal photo portraits. This is causing a shortage and, with the shortage, sky-high prices.

ALSO READ: This Vi Plan Under Rs 400 Is Built For Data-Heavy Users With 28-Day Validity and Unlimited Calls

Hassabis Believes There Is A Choke Point

DeepMind's CEO believes there could be a point when things get bottlenecked due to shortages in the supply of memory. Speaking to CNBC, he said, “You need a lot of chips to be able to experiment on new ideas at a big enough scale that you can actually see if they're going to work.” Hassabis describes this as a "choke point."

Zuckerberg has previously said that AI researchers want "the most chips possible."

At a time when AI firms are racing to build bigger models and push computational limits, the pressure on memory chip supply is emerging as a serious bottleneck for the entire sector.

Google is not immune to this crunch. While the company designs its own Tensor Processing Units, giving it some insulation from heavy dependence on external players like Nvidia, that advantage only goes so far. As Demis Hassabis pointed out, “It still, in the end, actually comes down to a few suppliers of a few key components.”

In fact, the strain has been significant enough that Hassabis acknowledged Google has been constrained to the extent that it has struggled to fully meet demand for its Gemini models.

Now as the shortage surges, the big tech giants like Google and Microsoft are turning their focus to South Korea to procure more chips. There are three main players in the market when it comes to chip production: Samsung, Micron, and SK Hynix, with Samsung and SK Hynix both positioned in South Korea. Micron Technology has decided to phase out its chip production for personal electronics and instead concentrate on manufacturing chips designed for artificial intelligence systems.

ALSO READ: Microsoft Names AI Executive Asha Sharma As New Xbox CEO, Signals Renewed Focus On Console Gamers

The semiconductor shortage is expected to persist, with little indication of near-term relief. Against this backdrop, Google has signaled a major escalation in its AI strategy, outlining projected capital expenditures of $175 billion to $185 billion in 2026 to expand and strengthen its AI infrastructure in anticipation of continued growth.