The speed and accuracy of quantum computing makes it ideal for collecting and analyzing data, modeling the impact of proposed tax legislation, and predicting and detecting tax evasion. Computers using quantum technology can instantly solve computational problems that could take an ordinary PC decades. These same qualities, however, threaten data security.
Quantum Computing
Quantum computing is based on quantum mechanics, or the study of the physical properties of atomic and subatomic particles. A quantum is the smallest physical unit of a system. For example, a quantum of light is a photon, and a quantum of electricity is an electron. The word quantum originates in Latin and means "an amount" or "how much?"
A quantum state in a subatomic system is a collection of physical properties that can be measured simultaneously and provides a probability distribution for the outcomes on the system of each possible measurement. Knowledge of the system’s quantum state and the rules for the system's evolution in time covers all that can be predicted about the system's behavior.
A quantum computer differs from a classical computer in its basic unit of information. The classical computer works on bits of 1 and 0, known as the dual-processing system. Quantum computers rely on quantum bits, or "qubits," which work with 1, 0, or a combination of both. Qubits can be engineered as photons, electrons, or atomic nuclei.
Qubits allow all information combinations to exist simultaneously in more than one place, in a phenomenon known as superposition. This allows quantum computers to calculate a multitude of equations or possibilities concurrently, a task done step by step by regular computers.
Qubits, like most subatomic particles, can become connected so that the action on one qubit can influence another qubit, in a phenomenon known as entanglement. If two qubits are entangled in the same quantum state, changing the state of one qubit will switch the state of the other one, even if the qubits are separated by enormous distances. This drastically increases the speed and accuracy of quantum computers, making them ideal for data collection, sharing, and modeling. Quantum computers can also facilitate the equalization, tracking, review, and correction of tax systems.
Speed and Accuracy
Since the EU introduced mutual administrative assistance in 2011, the exchange of tax information has expanded to include advance cross-border tax rulings, advance pricing agreements, country-by-country reporting, beneficial ownership information, cross-border tax arrangements, and sales on electronic platforms. Data collection and sharing are facilitated by systems that require public CbC reporting, the OECD’s pillars 1 and 2, and the common reporting standard.
These systems will succeed only if technology enables taxpayers to supply the data without breaking the system. Data sets that are too large or complex to be managed by traditional data-processing software are called big data. Big data can be collected and analyzed by combining quantum processors with artificial intelligence and improved machine learning.
Machine learning is mainly used for data classification. Quantum computers can create new classifiers that generate maps of more sophisticated data, enabling researchers to develop more effective AI that can identify patterns invisible to classical computers and classify the data more accurately.
Identifying and analyzing data is the goal of the OECD’s Analytical Database on Individual Multinationals and Affiliates (ADIMA). The ADIMA project examines the operations of 500 multinational companies to determine the location of their value chains, how they operate, and where they pay taxes.
The OECD plans to expand the database and include information not usually found in company reports by evaluating open big data sources like news outlets and websites. The first published reports reviewed only the top 100 ADIMA companies and found that 85 of them had active operations, while official financial statements indicated active operations for only 75.
Quantum technology can also provide robust tax impact analyses to governments before they enact legislation. Tax impact theory analyzes how changes in tax systems affect taxpayer response and is useful for examining cross-border models like the common consolidated corporate tax base. A quantum computer could simulate possible effects of proposed legislation on distribution of the tax base.
Fraud and Evasion
Most financial institutions invest in fraud detection systems that use advanced algorithms, but those systems can produce a high number of false positives, which drives organizations to be overly risk averse. Reviewing false alerts is time-consuming and can block legitimate transactions.
Data-modeling capabilities of quantum computers are superior in finding patterns, performing classifications, and making predictions. A support-vector machine sorts data into classes within a set of decision boundaries (called a hyperplane). Its algorithm learns by example to assign labels to objects. After the machines are trained, they can assign new data to the appropriate category.
Machine-learning techniques employed by quantum support-vector machines use supervised learning models to classify and regress data and to detect outliers, which reduces false positives.
Data Security
But there are drawbacks. Corporations, banks, and governments are aware of the data security risks of quantum computing. In 2015 the U.S. National Security Agency had quantum computing in mind when it warned intelligence agencies to choose cryptographic algorithms. In April IBM unveiled the z16 platform — promoted as the industry’s first quantum-safe system.
Quantum computing moves computers beyond binary logic into atomic-level randomness. Encryption that is mathematically complex enough for now could be decrypted by a quantum machine in the future. This would dramatically change the dynamics of fraud prevention. While the average citizen lacks the means and knowledge to acquire and use the technology (a commercial quantum computer with 2000 qubits is currently priced at $15 million), international cartels earning large revenues from ransomware could afford it.
Imagine a classic computer is attacked by an adversary with quantum technologies. Security can be breached retrospectively because encrypted messages can be intercepted, stored, and deciphered 10 years later.
This possibility has generated demand for quantum-safe cryptography. Post-quantum cryptography, or a type of cybersecurity that can be used by conventional computers, is under development. Quantum-resistant cryptography could prevent data exposure and improve protection of digital assets.
In 2016 the U.S. Department of Commerce’s National Institute of Standards and Technology launched a competition calling on academics and industry cryptographers to design an algorithm that could resist decryption from a quantum computer. On July 5 it chose four algorithms to include in its post-quantum cryptographic standard.