Google experts have unveiled a technology that, using artificial intelligence, can make quantum computing practical in real life. This is stated in an article in Nature.
In the presented work, DeepMind researchers explained that their new AI system AlphaQubit has been successful in correcting persistent errors that have long plagued quantum computers.
«Quantum computers have the potential to revolutionize drug discovery, materials development, and fundamental physics, if we can get them to work reliably,» states Google.
Super-powerful computing devices are extremely fragile: even the slightest environmental interference, such as heat, vibration, electromagnetic fields, or even cosmic rays, can disrupt quantum states. This will lead to errors and unreliable computations.
In one of the previously published works, it is stated that for practical application of quantum computers, the error rate should not exceed one in a trillion operations.
«Some tasks that an ordinary computer will take billions of years to solve, a quantum computer will only require a few hours. However, new processors are more susceptible to noise compared to ordinary ones. If we want to make quantum computers reliable, especially at scale, we need to accurately identify and correct these errors,» Google emphasized.
The new AI system AlphaQubit employs a complex neural network architecture that has demonstrated high accuracy in detecting and correcting quantum errors.
The solution maintained high accuracy in quantum systems ranging from 17 to 241 qubits. This suggests that the approach can scale to larger systems needed for practical quantum computing.
AlphaQubit faces significant hurdles before implementation.
«Each consistency check in a fast superconducting quantum processor is measured a million times a second. While AlphaQubit excels at accurately identifying errors, it is still too slow to correct errors in a superconducting processor in real-time,» claim the developers.
Researchers focused on optimizing speed, scalability, and integration.
The crypto community fears blockchain hacking by a quantum computer. Fears have intensified after the 'world's first effective attack' on Present, Gift-64, and Rectangle algorithms.
The fastest computer in the world
Although quantum computers have the potential to disrupt blockchains like the Bitcoin and Ethereum networks, this cannot be said about even the fastest classical supercomputers.
Scientists from the Lawrence Livermore National Laboratory announced that their newest supercomputer El Capitan can perform 2.79 quadrillion (a number with 15 zeros) calculations per second—making it the fastest computer in the world. It operates 5.4 million times faster than a standard home PC.
The device is intended for complex tasks such as modeling, AI development, and research. As for concerns regarding the likelihood of blockchain hacking, experts in cryptography believe they are unfounded.
«They need to check all possible private keys. For example, if the length of the private key is 256 bits, an attacker trying to forge your transactions would need to check all possible 256-bit keys,» noted Arcium's CEO Yannick Shreid.
El Capitan needs 10 billion years to do this due to the computational asymmetry inherent in the encryption schemes used, such as elliptic curve-based cryptography (ECC). It is used in various blockchains, including Bitcoin, Ethereum, and Solana.
ECC provides a high level of security, but a more serious threat comes from quantum computing, Shreid emphasized. They are capable of overcoming computational asymmetry.
«A supercomputer still processes data in binary form, using bits, and its power is primarily due to parallelization—simultaneously performing many tasks. Quantum computers, on the other hand, use quantum bits (qubits) that apply quantum principles such as superposition and entanglement. This is a different dimension of computation,» added Arcium's CEO.
Other experts expressed a similar position. Modern encryption systems cannot be hacked by any classical means. The threat posed by new quantum computers is much stronger, noted Duncan Jones, head of cybersecurity at Quantinuum. His company is working on quantum computing development.
«In just a decade, we will see sufficiently powerful quantum computers that can break blockchain algorithms. Developers of decentralized networks need to deploy quantum security technology as quickly as possible to prepare for such risks,» he noted.
Just do it
Quantum computers are still far off, and supercomputers cannot hack cryptocurrencies, but it is too early to relax. Professor Ahmed Banafa from San Jose State University noted that 'the blockchain industry is too trusting of its presumed security, overlooking potential vulnerabilities.'
«Many advocates consider the immutability of the blockchain synonymous with invulnerability, but such a viewpoint ignores the risks associated with the development of technologies like quantum computing, and practical issues such as software flaws,» he stated.
While crypto users acknowledge the potential threat, only a few platforms have taken measures to protect against such attacks, the expert added.
«Decentralization is a strong point, but at the same time, it complicates the process of deploying critical updates to enhance security,» emphasized Banafa.
Previously, several experts stated that the threat of a quantum attack on cryptocurrencies is exaggerated.
#новости #newsdaily #newscrypto #BTC☀ $BTC