Our approach to ethics and how it guides our work

When your work directly impacts people and societies, ethics matters. This is how we ensure our projects are ethically sound.

At Faculty, we believe AI is a force for good. We are privileged to work on projects that benefit not just our customers, but wider society, and care deeply about our reputation for working with integrity.  

You can see that in how we helped the NHS forecast covid admissions and save thousands of lives. You can see how we have helped improve the standards of online advertising. You can see how we have helped the London Fire Brigade target their inspections more efficiently. You can also see how we helped reduce train delays, or helped government combat terrorist propaganda online. 

We are fortunate to support our customers in such impactful work and, as a result of our purpose-driven culture, we think carefully before deciding whether to undertake a project. As such, our leadership team – and indeed the whole company – are responsible for reflecting on and verifying the ethical basis of a project before we decide to take it on. 

We do this by being very selective about our customers in the first place, applying our ethical principles to the project and – if necessary – by referring a potential project to our Ethics Panel. In addition to always acting with integrity and honesty, we apply our ethical principles to each project.

Our principles

We adhere to the following principles in all of our work:

1.No one should have to do work that they believe is unjust or makes them uncomfortable.

FAI offers solutions to a range of partners across various sectors, some of which you may feel uncomfortable about working on. We respect your autonomy and conscience, valuain a collaborative community of equals working to improve the world. Employees have the freedom to decline to work on projects or sectors they are uncomfortable with or believe to be unjust. Justice has been the subject of disagreement among reasonable people throughout the history of reflection on ethics. FAI’s principles split the difference: we do not take a strong stance on what justice means, but grant employees the benefit of the doubt to accommodate their conscience and personal ethical standards.

2. We will never risk our scientific credibility.

We believe in the scientific method and maintain our own professional standards as legitimate and credible scientists in all of our work. We adhere to established and sound scientific methods and principles and will not sacrifice our scientific standards or credibility to a customer’s requests or demands.

3. We don't do work that has the aim or effect of unjustly or illegitimately exploiting vulnerability.

Exploiting someone means taking advantage of their vulnerabilities, such as ignorance or disability. However, it can be just and legitimate to exploit vulnerabilities in some cases, particularly in defence or law enforcement, to achieve just aims against a liable target. We must distinguish between morally acceptable and unacceptable exploitation. Some targets, like enemy combatants or suspected criminals, may be justly liable and thus not wrong by exploitation. We defer to our defence and law enforcement partners to make these distinctions in good faith, within the bounds of legal and moral safeguards such as the law of armed conflict, international humanitarian law, and various local laws protecting individual rights in law enforcement.

4. We respect the letter and the spirit of the law.

Ethics often requires more than legal compliance, especially when the law is unclear, underdeveloped, or contains loopholes that conflict with its intent. In these cases, we follow the spirit of the law when this is plain and uphold key principles of justice, humanity, accountability, and due care.

5. We don't do party politics (but will support Al policy).

We do not endorse, or lobby for political parties or partisan political causes. However, within the field of AI, we may support legislation, research, and policy development that align with our ethical principles, and generate positive societal impact, such as promoting safe and effective AI for society and communities.

6. We take due care in choosing customers to work with, especially when working outside the UK, and apply a selection framework to minimise risk.

We have a responsibility to work with customers who will use our technology responsibly. We also recognize that exporting our technology outside of the UK raises especially sensitive concerns about enabling foreign regimes and their behavior. In these cases, we apply a tiered approach to which countries and regimes we will work with. We are developing a matured “customer selection framework” to mitigate the potential harms of technology transfer and provide a more consistent objective basis for these decisions.

7. We pick our own teams

We do not alter our team compositions based on customer preferences or pressures. We select our teams based on merit and inclusion, ensuring no team members feel undue pressure to participate in any project.

Our Ethics Panel

Our Ethics Panel was set up to support our teams in the consistent and fair application of our ethical principles. The panel consists of seven randomly appointed employees from across Faculty, including three senior and four more junior members of staff, and with panel membership renewed every six months.The panel judges whether a project is aligned with the principles set out above, whether there is a possible conflict, or whether a project raises ethical issues not covered by our principles, but which should be. 

Every project offer Faculty receives is tested against our ethical principles by our people before a contract is signed. It is a non-negotiable and an established part of our preparatory work. 

This is an evolving process that we are continually reviewing, and there will always be more we can do to ensure all our work is as sound and ethical as possible.

But our governance practices should leave our customers and employees in no doubt that we hold these matters in high regard, have no wish to be involved in ethically questionable projects, and fundamentally believe our work should only be a force for good.