How IBM Is Making Your Passwords Useless
For years, quantum computing has been hailed as a technology that could change the way the modern world works, but a long-standing technical issue has kept that potential from being realized. Now, in a paper published in the journal Nature last week, IBM scientists have taken a big step (see how I avoided the temptation to make a pun there?) toward solving that problem — and while it could represent progress toward making quatum computers real, it also could mean that current cybersecurity standards will soon be much easier to crack. In other words, your passwords could be obsolete soon.
The power of quantum computing has some obvious appeal: The increase in processing power could speed up research, especially in big data applications. Problems with large datasets, or those that need many millions (or billions, or more) of simulations to develop a working theory, would be able to be run at speeds unthinkable today. This could mean giant leaps forward in medical research, where enhanced simulations can be used to test cancer treatments or work on the development of new vaccines for ebola, HIV, malaria and the other diseases. High-level physics labs like CERN could use the extra power to increase our understanding of the way the universe at large works.
But the most immediate impact for the regular person would be in the way your private information is kept safe. Current encryption relies on massively large prime numbers to encode your sensitive information. Using combinations of large prime numbers means that anyone trying to crack such encryption needs to attempt to factor at least one of those numbers to get into encrypted data. When you buy something from, say, Amazon, the connection between your computer and Amazon is encrypted using that basic system (it's more complicated than that, but that's the rough summary). The time it would take a digital computer to calculate these factors is essentially past the heat death of the universe. (Still, this won't help you if your password is password, or monkey, or 123456. Please, people, use a password manager.)
Quantum computing, however, increases processing speed and the actual nature of the computation so significantly that it reduces that time to nearly nothing, making current encryption much less secure.
The IBM researcher that could make that happen is complicated, and it requires some background explanation. For starters, while a "traditional" computing bit can be either a 0 or a 1, a quantum computing bit can have three (or infinite, depending on how you want to interpret the concept) states. More specifically, a qubit can be 0, 1, or both.
Up until now, the both part of that caused some problems in realizing the power of quantum computing.
Apparently — and you'll have to take this on faith a bit, as it hurts my head to think about it — the both state can switch back to either 0 or 1 at any given point, and sometimes incorrectly, based on the logic in the programming. Think about when your phone freezes up for a second or two while you're matching tiles. This is its processor handling vast amounts of information and filtering out the operations that fail for any number of reasons, from buggy code to malware to basic electrical noise. When there are only the two binary states, this is a process that usually happens behind the scenes and quickly.
The hold-up with quantum computing up until now is that the vastly greater potential for errors has stymied attempts to identify and nullify them. One additional wrinkle in this reading quantum states is familiar to anyone with basic science fiction knowledge, or perhaps just the ailurophobics. What if the action of reading the qubit actually causes it to collapse to 0 or 1?
The very smart people at IBM think they've solved this. The actual technical explanation is involved, and well beyond my ability to fully follow, but the gist is that instead of just having the qubits arrayed in a lattice on their own, they are arranged such that neighbors essentially check each other, producing the ability to check the common read problems.
That opens the door to further quantum computing developments, including ones that will make your password a thing of the past. So, does this mean that you need to start hoarding gold? No, not yet. And hopefully before quantum computing reaches commercial, or even simply industrial/governmental levels, a better cyber security method will be in place. Or the robots will have already taken over. I for one welcome them.
Number of the Day: 51%
More than half of registered voters polled by Morning Consult and Politico said they support work requirements for Medicaid recipients. Thirty-seven percent oppose such eligibility rules.
Martin Feldstein Is Optimistic About Tax Cuts, and Long-Term Deficits
In a new piece published at Project Syndicate, the conservative economist, who led President Reagan’s Council of Economic Advisers from 1982 to 1984, writes that pro-growth tax individual and corporate reform will get done — and that any resulting spike in the budget deficit will be temporary:
“Although the net tax changes may widen the budget deficit in the short term, the incentive effects of lower tax rates and the increased accumulation of capital will mean faster economic growth and higher real incomes, both of which will cause rising taxable incomes and lower long-term deficits.”
Doing tax reform through reconciliation — allowing it to be passed by a simple majority in the Senate, as long as it doesn’t add to the deficit after 10 years — is another key. “By designing the tax and spending rules accordingly and phasing in future revenue increases, the Republicans can achieve the needed long-term surpluses,” Feldstein argues.
Of course, the big questions remain whether tax and spending changes are really designed as Feldstein describes — and whether “future revenue increases” ever come to fruition. Otherwise, those “long-term surpluses” Feldstein says we need won’t ever materialize.
JP Morgan: Don’t Expect Tax Reform This Year
Gary Cohn, President Trump’s top economic adviser, seems pretty confident that Congress can produce a tax bill in a hurry. He told the Financial Times (paywall) last week that the Ways and Means Committee should be write a bill “in the next three of four weeks.” But most experts doubt that such a complicated undertaking can be accomplished so quickly. In a note to clients this week, J.P. Morgan analysts said they don’t expect to see a tax bill passed until mid-2018, following months of political wrangling:
“There will likely be months of committee hearings, lobbying by affected groups, and behind-the-scenes horse trading before final tax legislation emerges. Our baseline forecast continues to pencil in a modest, temporary, deficit-financed tax cut to be passed in 2Q2018 through the reconciliation process, avoiding the need to attract 60 votes in the Senate.”
Trump Still Has No Tax Reform Plan to Pitch
Bloomberg’s Sahil Kapur writes that, even as President Trump prepares to push tax reform thus week, basic questions about the plan have no answers: “Will the changes be permanent or temporary? How will individual tax brackets be set? What rate will corporations and small businesses pay?”
“They’re nowhere. They’re just nowhere,” Henrietta Treyz, a tax analyst with Veda Partners and former Senate tax staffer, tells Kapur. “I see them putting these ideas out as though they’re making progress, but they are the same regurgitated ideas we’ve been talking about for 20 years that have never gotten past the white-paper stage.”
The Fiscal Times Newsletter - August 28, 2017
|