Emerging technologies and the rule of law
04 November 2021
Quantum will enhance emerging technologies, such as, Artificial Intelligence. Their relationship will supercharge legal issues that have so far arisen with AI and will of course create new ones. What role will lawyers play in this new quantum charged world? How can lawyers protect the rights of individuals in this quantum simulated metaverse?
With qubits existing in multiple states simultaneously, quantum phenomena such as superposition and entanglement allow for unprecedented computational and predictive power. This has the potential to solve complex problems – for vast amounts of data to be digested and complex variables examined to perform predictive analysis at speeds unheard of today which can then form the basis for machine learning and automated decision making.
However, for all the potential this has, data used to train the quantum enhanced AI may suffer from societal and historical data bias which could then result in discrimination and unfair decisions. Automated decisions based on biased data creates prejudiced decisions with affected individuals bearing the onus of having to prove that the effect of the algorithms on them is contrary to the law. As automated systems often do not provide an opportunity for the discretion of a human decision maker to be exercised, the affected person is usually not able to present their defense until after a decision is made. Essentially, they are guilty until they can prove they are innocent. The simple fact is that fixed immutable rules cannot always deal with the particularities of an individual's circumstances. Even with quantum computing able to analyse vast quantities of variables and potential outcomes, a person is entitled to due process, to make representations concerning a decision that affects them. They are entitled to discretion, equity and compassion, which is where lawyers come in – to sort out the facts from the qubits.
Examples today (even without the quantum boost) include COMPAS which was designed to evaluate the risk of criminals re-offending but was found to discriminate against black people, and a MIT Media Lab researcher testing Microsoft, IBM and Megvii found that face-analysing AI works significantly better for white faces than black ones. These illustrate the areas of discrimination law, criminal law, employment law and tortious actions which will constitute novel factual complexity likely involving computer code that lawyers will have to untangle. Who should be liable, the owners of the hardware, the software manufacturers, the parties setting the parameters and the data sources, the coders, or the users? What rights do we have to protect or delete our digital twin (our digital data that can be mined from social media, medical records and employment records etc.). And this does not even begin to touch on the legal issues that will arise when neural networks are developed with the aid of quantum to the extent that we can create autonomous beings. Criminal law requires the meeting of mens rea (a guilty mind) with the actus rea (criminal act). Can an autonomous machine have a guilty mind?
Fundamentally, lawyers have a duty to the administration of justice, to uphold the rule of law. Lawyers are obliged to protect the right of an individual to equality before the law and to oppose the influence of arbitrary power. In a quantum enhanced AI world, the role of lawyers will be crucial in ensuring that quantum does not heighten the biasness of history but embraces fairness, transparency and accountability.
Quantum should reflect core societal values (currently enshrined in our legal system) and with regards to quantum enhanced AI, there are five main areas where we see lawyers playing a significant role:
- protecting the rights of individuals (the right to due process, anti-discrimination laws, employment law, criminal law, privacy and data security);
- ensuring operational transparency to ensure that fairness metrics are maintained in accordance with legal standards;
- determining liability at various points in the AI quantum chain (who will be liable, when will they be liable and for what will they be held liable);
- creating ethical standards for regulating the research and application of quantum technologies; and
- creating international standards on access to quantum technology.
This article was written as part of techUK's Quantum Commercialisation week.