Think AI Is Baffling? Here's How to Pretend You Understand Quantum Computing. -- Barrons.com

Dow Jones
2024-12-14

By Jack Hough

What an embarrassing week to be a supercomputer. These are machines that connect many processors to work on complicated jobs, like modeling weather or conjuring a koala-in-a-tophat emoji. Going fast is kind of their whole thing. But a software company just dunked on the entire industry. Google said that it has a new "quantum computing" chip called Willow that can perform a calculation in less than five minutes that would take today's fastest supercomputers longer than the age of the universe.

"Our logical qubits now operate below the critical quantum error correction threshold," explained a company scientist in a windbreaker and sneakers who I'm pretty sure can bend metal with his mind. It's a significant development in terms of chained geometric equity progression, meaning that whatever that guy was talking about made shares of parent Alphabet go up more than 5% on both Tuesday and Wednesday.

My first thought was whether quantum computing's potential for good in fields like drug discovery will outweigh its dangers in code breaking and cyberwarfare. But my even first-er thought was whether there's an exchange-traded fund. It turns out that the Defiance Quantum ETF (ticker: QTUM) is up 38% this year, versus 29% for the Global X Artificial Intelligence & Technology ETF $(AIQ.UK)$, meaning that quantum computing is suddenly even hotter than AI.

What's that, you say? Quantum computing and AI aren't alternatives but rather complements, and to suggest otherwise raises questions about my quantum computing bona fides? Stop trying to blow my cover. Quantum computing is already popping up in water cooler conversations across ordinary offices, which means you need to understand it in a hurry -- only I've tried, and the stuff is impenetrable. What you really need is a guide to pretending to understand quantum computing. And there I think I can help.

Step 1 is to explain to no doubt impressed onlookers that classical computing is based on bits, which are stored in two possible states, 0 or 1. "That's why we call them bits," be sure to say. "It's a portmanteau for binary digits." Pause here for admiring nods.

Next, explain that quantum computing is based on what are called qubits. "That's with a q," make clear, or else your colleagues will picture cubits, and Noah's Ark -- we're going in a totally different direction. A qubit is a unit of measurement based on what's called a superposition of multiple possible states. It's difficult to understand this, because it comes from quantum mechanics, or how subatomic particles behave, which is totally different from how the objects around us behave.

In quantum mechanics, particles can be in two states at the same time until they're observed. There was an exasperated Austrian called Schrödinger who tried to explain this to normals in terms of a make-believe cat trapped in a box with an unstable radioactive material that may or may not decay, triggering a Geiger counter, which releases poison, such that you're not sure whether the cat is dead until you open the box, or as quantum mechanics would have it, that the cat is both alive and dead, in a superposition of both, with a certain probability of each. For me, this only raises more questions, including whether Schrödinger had pets.

Best to skip superposition when explaining quantum computing, but remember the term, along with entanglement, decoherence, and interference. If your colleagues chime in with any of these, it means they know more than you. Immediately fake a sneezing fit and leave the conversation.

The upshot is that classical, bit-based computers are deterministic, solving problems by trying solutions one by one, sequentially, which means they can get jammed up like the midday Christmas break standby line for Disney World's Jungle Cruise. But quantum computers are probabilistic and massively parallel, which you might say makes their thinking more human, except for the part about running on a ball of wrinkled fat in a skull.

Don't give your audience more than they can handle. Wrap up by explaining that since subatomic behavior forms the foundation of the universe, all systems are effectively built on quantum systems, including traditional computers -- they just don't yet use quantum properties in their calculations. As you say this, gaze distantly upward to appear extra-insightful. Definitely don't mention that you got it from an IBM blog -- it kills the mood.

One investment approach here is, apparently, to just buy anything with the word quantum in it. Quantum Computing is up 584% year to date; D-Wave Quantum, 344%; and Quantum Corp., 154%. But that strategy risks missing popular players with disadvantaged names. Rigetti Computing is up 547% this year, and IonQ, 148%. As far as where we are in the cycle, it feels at least five years too soon for the free cash flows, and about five weeks too late for the easy stock gains -- but who knows?

Google's breakthrough with Willow, by the way, is to exponentially reduce errors as the system expands, which is the opposite of what typically happens. Commercial uses could be years away, but quantum innovation is already building a significant technology moat for Alphabet, which isn't reflected in the stock's valuation, reckons BofA Securities analyst Justin Post.

In five to 10 years, the most demanding computing might already be done on quantum machines, predicted D.A. Davidson at the start of the year. That could make Nvidia "vulnerable," the firm wrote. But the stock is up 179% this year. Computer scientists say quantum systems are likely to be used only for some jobs, in a hybrid arrangement with traditional computers. Nvidia has been announcing partnerships that use its chips to make quantum breakthroughs.

I still can't figure out whether Schrödinger's cat is still alive, but I won't be surprised if Nvidia makes a killing from it.

email: jack.hough@barrons.com

This content was created by Barron's, which is operated by Dow Jones & Co. Barron's is published independently from Dow Jones Newswires and The Wall Street Journal.

 

(END) Dow Jones Newswires

December 13, 2024 15:22 ET (20:22 GMT)

Copyright (c) 2024 Dow Jones & Company, Inc.

免责声明:投资有风险,本文并非投资建议,以上内容不应被视为任何金融产品的购买或出售要约、建议或邀请,作者或其他用户的任何相关讨论、评论或帖子也不应被视为此类内容。本文仅供一般参考,不考虑您的个人投资目标、财务状况或需求。TTM对信息的准确性和完整性不承担任何责任或保证,投资者应自行研究并在投资前寻求专业建议。

热议股票

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10