Operationalising Ethics
Trust is at an all time low. This matters not just because trust disproportionately impacts bottom line business outcomes, but also because the relationship between data sharing and trust is becoming clearer. People want and expect control of their data. It’s no longer accepted that limited value is assigned to privacy. There is a tradeoff fallacy.
Our societies and economies are increasingly information driven. Information asymmetry, power imbalances and monopolies are threatening our ways of living. Individuals face increasing risk of manipulation, along with a variety of other harms that aren’t yet well understood. The ineffectiveness of the current model, particularly the lack of active participation from individuals, is resulting in billions of unrealised economic benefit.
Ethics — specifically ‘Data Ethics’ — has never been more important to modern information businesses. The time is now to take a leadership position. The time is now to become ethical by design.
Data Ethics is an emerging field of study. It focuses on evaluating moral challenges related to how information is used in order to help develop morally ‘good’ solutions.
Much work is being done to establish principles. Accenture, Google and Microsoft are high profile examples of big tech companies making moves. The UK Government has invested heavily as part of their National Data Strategy, establishing principles and supporting workbooks, along with the Centre for Data Ethics and Innovation. The Open Data Institute (ODI) continues to evolve their Data Ethics Canvas. The European Commission has done considerable work to collaboratively develop ‘Ethics guidelines for trustworthy AI’. The Gradient Institute was formed to research the ‘Science of Ethical AI’.
Other approaches, like DEON, a command line tool that support ‘ethics checklists’ is increasing in popularity. Data for Democracy’s, ‘The Global Data Ethics Project’ has received support from thousands of practitioners. Thoughtful work is being produced by The Ethics Centre and the IEEE.
Although much progress is being made, new products and services are being developed faster than ever before. The market continues to make mistakes, reinforcing biases and unintended uses that result in material and immaterial harms.
We’d argue that organisations need a framework.
A Data Ethics Framework is the consistent process an organisation executes to decide, document and verify that its data processing activities are socially preferable.
Social preferability, rather than acceptability, is the highest order value.
Although organisational context matters, this diagrammatic representation of a Data Ethics Framework has been designed to help you develop an understanding of the specific functions — and relationship between functions — that can assist you in progressively operationalising socially preferable data enabled business activities.
Let’s get started.
It’s broadly accepted that boards are responsible for corporate governance. And although there doesn’t seem to be a broadly agreeable definition, it’s accepted that corporate governance is focused on setting strategy and identifying and mitigating risk. Given that most businesses today are information businesses by design, it’s arguable that Information Governance should be a board level responsibility. Boards should contribute to setting ambitious information strategy, whilst effectively identifying and managing information risks.
Doing this helps create the authorising environment for data ethics to be taken seriously across the business.
In this context, the strategic intent to operationalise a Data Ethics Framework becomes part of an organisations strategy. Committing to this creates new uncertainties, or rather, risks to be identified and mitigated.
If you’d like to dive deeper into this specifically, get in touch.
Moving on, we’d argue this strategic intent can become a key component of a Trust by Design or trustworthiness strategy. This can be developed by asking and answering 5 key questions:
- Where are we today?
- Where would be like to be in the future?
- What will we do to get there?
- How will we do it? And
- How will we know if we’re making progress?
Again, for more information on how to design and execute a trustworthiness strategy, reach out.
From here there are really four specific functions worth calling out:
- External Data Ethics Committee
- Stakeholder collaboration
- Independent audits, and
- Change programs
An external committee will bring new, independent expertise into the organisations view. More specifically, it provides a more objective opportunity to critically assess intent and outcomes from different viewpoints.
Stakeholder collaboration, specifically, Social Preferability Testing, provides a mechanism through which the intent and outcomes of proposed (or existing) data processing activities can be put to the test with the group of stakeholders participating in the ecosystem an organisation contributes to. After all, just because an organisation states something is ethical doesn’t make it so. Organisation should be optimising their activities for overwhelming support. Regulators, independent advocacy groups and perhaps most importantly, customers, should be ecstatic about what an organisation intends and actually does with data.
For more on how to build this body of evidence systematically, get in touch.
Independent audits are an opportunity to assess intent, tradeoff decisions, evidence established to date, and situate that in a broader context. Will the good outweigh the bad? What could be done to mitigate risk? Where haven’t customers been engaged in the process? What specific insight might we transfer to this proposal to increase the confidence with which we move forward?
This is also another mechanism through which organisations can be held accountable for their actions. After all, the tech we build is never neutral, regardless of what technological instrumentalists say.
If this is really piquing your interest, it’s worthwhile diving into other resources, like The Ethics Centre’s Principles for Good Tech.
Lastly it’s worth discussing behavioural change. After all, if we do not change the way we execute workflows on a daily basis, we cannot hope to operationalise and effectively evidence a Data Ethics Framework. A DEF requires all hands on deck. It’s a cultural and operational statement to fight for what you believe is right.
Going back to the very start, if we assume trust is a proxy for ethics (as philosophers like Dr Matt Beard argue), then an effective Data Ethics Framework can act as a foundation to earn and be worthy of consumer trust. This earned trust will help an organisation gain access to more of the right data. This access to the right data can be used to develop valuable, meaningful and engaging lifestyle services. The organisations delivering these lifestyle services will take a leadership position in their market.
If you’d like to discuss any of this in more depth, reach out to us through our website on greaterthanexperience.design
Or, if you’d like to dive deep into the broad considerations that might impact our ability to collaboratively design an empowering, prosperous and humanity centric future, get your hands on our most recent report.
