There’s an old saying in the semiconductor world: chips are the new oil. That metaphor, like most good ones, has limits. Oil runs downhill. Chips run on demand. And sometimes, as in the case of Samsung’s stunning recent slump, the demand disappears just when everyone thought it would never stop.
So, let’s walk this one through. Classic journalism style. Who, what, when, where, and why. And while we’re at it, how a company that defines the memory chip industry found itself playing catch-up in the age of AI.
Who: Samsung’s Crown Jewel and Its Global Rivals
Samsung Electronics isn’t merely a smartphone behemoth. It’s the largest memory chipmaker on Earth. Its semiconductor arm was, until recently, the profit powerhouse behind the entire Samsung empire. At its peak, the division contributed half of the company’s profits. That is no rounding error.
The competition is fierce and familiar. SK Hynix, Micron, and TSMC make up the rest of the global oligarchy. Yet Samsung’s size, once an advantage, became an albatross when the tide turned. While TSMC leaned on long-term contracts and SK Hynix leaned into high-bandwidth memory for AI, Samsung bet big on scale and staying the course. That bet didn’t age well.
What: From Record Profits to Historic Losses
Samsung’s chip division posted an operating profit of 9.98 trillion won in Q2 2022. One year later, it reported a 4.36 trillion won loss. The first half of 2023 saw losses exceeding $7 billion. The collapse wasn’t a slow decline. It was an air-pocket drop. For a business once described as “too big to falter,” this was a staggering reversal.
Executives didn’t sugarcoat it. “A shock,” one said. Analysts spoke of crisis. And the headlines, when they came, sounded like obituaries for a business model. The company that built its reputation on resilience was now booking its lowest operating profit in 14 years.
When: The Downturn That Caught Everyone Off Guard
The seeds were sown in late 2022. After a pandemic-fueled chip buying spree, demand from PC and smartphone makers evaporated almost overnight. By October, Samsung was reporting its first profit declines in three years.
Competitors responded quickly. They cut production. Samsung didn’t. The company stuck to its guns, flooding the market with chips nobody wanted. Prices plummeted by 70 percent across several memory categories. It wasn’t until April 2023 that Samsung finally blinked and slashed output, an extraordinary move for a firm known for riding out downturns without flinching.
Even today, in mid-2025, Samsung’s chip unit is operating with profits down more than 90 percent from their peak.
Where: The Cracks in the Global Map
Samsung’s missteps were exacerbated by geopolitics. U.S. export controls blocked the company from delivering advanced chips to Chinese clients. AI chips built for leading customers sat idle in warehouses. At the same time, Chinese competitors ramped up production of lower-end chips, creating more price pressure on the segments where Samsung still played.
The company’s Texas foundry, which was supposed to signal a new era of manufacturing dominance, ran into delays. Meanwhile, SK Hynix became the sole supplier of high-end HBM to NVIDIA, today’s hottest chip customer. Samsung had the fabs, but not the deals. And in semiconductors, a half-step behind might as well be a mile.
Why: A Perfect Storm of Miscalculations and Market Forces
Samsung got several things wrong, some tactical, others structural. The company stayed too long in commodity DRAM and NAND while the real action shifted to specialized AI memory. It invested heavily in foundry capacity but couldn’t match TSMC on yield or trust. And it kept producing chips even as prices cratered, hoping that its scale would once again rescue the bottom line.
To be fair, not everything was under Samsung’s control. The macro picture was brutal. Rising interest rates, consumer pullback, and a general electronics slowdown created headwinds for every player in the space. But the lesson here isn’t about macroeconomics. It’s about strategic agility.
How: Rebuilding in a New Semiconductor World
Samsung isn’t dead. Far from it. Its market share in memory remains dominant. Its deep pockets and diversified portfolio (phones, TVs, appliances) will keep the lights on. It just signed a $16.5 billion deal with Tesla to supply AI chips. And it is ramping up HBM production, hoping to claw back share in the hottest segment of the industry.
Still, one can’t help but wonder whether this stumble will mark a pivot point. The chip industry is moving fast, pulled forward by artificial intelligence and geopolitics. The next chapter won’t be written by who has the most fabs, but by who has the fastest read on what comes next.
The Takeaway
In the end, this story isn’t about one company. It’s about how even the strongest players can get caught flat-footed in a market that rewards foresight more than force. For Samsung, the climb back up won’t just be about making chips. It will be about proving that it still knows how to play the game.
What do you think? In rapidly evolving tech markets, is scale still an advantage, or has agility become the new superpower?