Cryptography experts are betting thousands of dollars on how long it will be until quantum computers can render modern encryption methods obsolete 鈥� but the smart money seems to be on this being a few decades away.
Writing on a 鈥� a field that studies encryption methods that are resistant to the increased power of quantum computers 鈥� at Ericsson Research in Sweden offered a wager of $2050 that no quantum machine would break a 2048-bit RSA key by 2050.
RSA is a gold-standard form of encryption that uses keys created from prime numbers to 鈥渓ock鈥� away data to keep it safe, while the 2048 bits describes the size of such a key. A key of this size was the largest included in a to see if anyone could break this encryption, although the $200,000 prize money was never claimed. Longer keys require more computing power to crack, and are therefore more secure.
Advertisement
But while classical computers have struggled to do this with longer RSA keys, quantum computers can exploit the unusual properties of quantum physics to speed up the necessary calculations. They will render current encryption techniques obsolete once the hardware is sufficiently powerful and accurate. Exactly when that will be is open to debate.
Mattsson says that his motivation for making the wager was to spark discussion about the likely timescale, as many quantum computing start-ups are making bold claims about progress that he places 鈥渮ero trust鈥� in.
鈥淚 think there鈥檚 so much hype and so many false claims from the quantum community,鈥� he says. 鈥淪o I felt like somebody needs to say 鈥榥o鈥� to this. I think we will see a聽惫别谤测听drastic reduction in the quantum computing industry.鈥�
That said, Mattsson does think some organisations, such as national security agencies, do need to urgently move to post-quantum cryptography algorithms, otherwise adversaries could collect data now to decrypt in future decades when quantum computers are powerful enough. 鈥淭hey need to keep information secret for 50, 75 years,鈥� he says.
Mattsson鈥檚 wager led to a queue of cryptography experts keen to stake their own cash. One was , who took the US government to court in the 1990s to get it to allow the export of strong encryption software. He took the bet and gave his prediction for cracking the key as 2032.
Bernstein also says that spy agencies saying such a moment is 鈥渄ecades away鈥� might be trying to mislead the public and slow the roll-out of quantum-resistant encryption algorithms so that they can continue to crack messages. 鈥淓very day that we fail to act is giving away more user data to attackers,鈥� he says.
, who works with technical standards body the Internet Engineering Task Force investigating post-quantum cryptography, says he didn鈥檛 accept the bet because he鈥檚 65 and 鈥渦nlikely to care about the outcome 20 years from now鈥�. But he says that quantum computers will crack encryption in 2060 鈥渁t the very earliest鈥�, and that it is entirely possible that the moment will never come.
