Claude, Gemini and ChatGPT — three large models built by three rival labs — were each handed the same question. A Wall Street analyst has put an 850,000-won target on Samsung Electronics. What to make of it?
What stands out is not that the three reached wildly different answers. It is that, reasoning along separate paths, they arrived at a strikingly similar place. The question was never simply whether Samsung's share price would rise or fall. It was where Samsung will stand in the age of AI.
The number was real. Mehdi Hosseini, a senior analyst at the U.S. investment firm Susquehanna, set a target of 850,000 won on Samsung Electronics, a call that surfaced via Bloomberg in early June and drew global attention. Samsung was trading near 349,000 won at the time. The figure implied more than a doubling — and it landed while most domestic brokerages were clustered in the 500,000-to-600,000-won range. The basis for the target had not been fully disclosed, which only sharpened the debate.
But the number mattered less than the logic behind it.
Three paths up the same mountain
Claude fixed on Samsung's sheer production capacity. SK hynix may lead the HBM market today, the argument ran, but the eventual winner of the AI era is whoever can supply vast quantities of memory reliably. Technology can be chased; capacity cannot be conjured overnight. Samsung's manufacturing footprint, among the largest on earth, remains a formidable weapon.
Gemini zoomed out from the single company to the whole cycle. Data-center construction is racing ahead across the U.S., China, Europe, Japan and the Middle East at once. Just as steel boomed when the railways were laid and telecoms boomed when the internet spread, the reasoning went, the proliferation of AI should lift the entire memory industry.
ChatGPT framed it as a contest for AI supremacy. Many treat Nvidia as the center of the revolution — not wrong, but incomplete. GPUs alone do not make AI work. They need HBM, DRAM and SSDs, and the firm able to supply most of that is Samsung. On this reading, 850,000 won is not a price on today's earnings but a price on tomorrow's AI hegemony.
Three different routes, one summit: the ultimate winner of the AI build-out is not merely whoever makes the chip, but whoever supplies the infrastructure.
What the number actually means
People tend to read a stock price as a fact about the present. History suggests it is usually a bet on the future.
Microsoft in the 1980s, Amazon in the 1990s, Apple in the 2000s — in each case the market moved before the consensus did, pricing in a world that did not yet exist. The 850,000-won case for Samsung belongs to the same family. It is not a verdict on the company as it is, but an act of imagination about Samsung in 2030, in 2035.
If Samsung captures the HBM4 market, restores its foundry competitiveness and establishes itself as a core supplier to AI data centers and physical AI, 850,000 won is not an impossible number. If those conditions fail, it is a mirage. The number is only an outcome. What matters is the force that produces it.
History has always defied common sense
It is easy to forget that, barely 40 years ago, Korea's chip industry was no match for Japan's. NEC, Toshiba, Hitachi and Fujitsu ruled, and even the U.S. struggled against the Japanese advance. Few imagined a Korean firm would one day dominate global memory.
Samsung did it anyway. Chairman Lee Kun-hee's instruction to "change everything except your wife and children" was not a slogan but an obsession with quality. Samsung overtook Japan and became number one. History tends to move that way — the impossible becomes real, and the obvious collapses. When Wall Street floats 850,000 won, it is not naive optimism; it is memory of what Samsung has done before.
What Samsung still has to do
Sober assessment leaves a long to-do list.
First, HBM. If data is the crude oil of the AI era, HBM is the engine that refines it, and SK hynix holds the edge today. Samsung must prove itself in HBM4 to justify the bull case.
Second, foundry. TSMC is already the world's strongest. Catching up demands not only technology but the return of customer trust — winning back the likes of Qualcomm, Nvidia, AMD and Apple.
Third, talent. The chip war is, in the end, a talent war, and the AI era rewards engineers who span design, packaging, software and algorithms.
Fourth, physical AI. Until now AI lived on a screen. Next it becomes the robot, the autonomous vehicle, the smart factory — and demand for silicon explodes. To be a true winner, Samsung must evolve from a memory company into an AI hardware platform.
Templeton's warning, Buffett's plain truth
Markets always carry risk. The investor John Templeton is widely credited with the observation that bull markets are born on pessimism, grow on skepticism, mature on optimism and die on euphoria. People believed in permanent prosperity before 1929, in eternal internet growth before 2000, in ever-rising home prices before 2008. The cycle repeated each time. Templeton's point was to stay coldest when the market runs hottest.
Warren Buffett tells a simpler version of the same story: invest in businesses you understand, with a durable advantage and able management, at a sensible price — and be fearful when others are greedy, greedy when others are fearful. The lesson is not to be swept along by the crowd. For anyone weighing the 850,000-won case, the point is the same. The number is not what counts. Whether Samsung is genuinely recovering its edge in HBM, foundry and AI infrastructure — that is the heart of it.
A story about people buying the future
The 850,000-won case is not really a stock story. It is a question about where Korea will stand in the AI revolution. The three chatbots analyzed the future each in their own way; Wall Street offered a number; investors talk in dreams. In the end, though, the market is settled by the cold realities of technology, earnings and innovation. Technology builds hope, management realizes it, results prove it — and the share price follows behind.
So the real meaning of Samsung's 850,000-won case does not lie in a price target. It lies in a single question: can Korea's flagship company, once again, challenge for the summit at the center of an industrial revolution? The answer will be written not by Wall Street, and not by AI, but by Samsung itself — in the history of technology and innovation it has yet to write.
Copyright ⓒ Aju Press All rights reserved.
