Social Licence and Digital Trust in Data-Driven Applications and AI: a Problem Statement and Possible Solutions

Abstract

‘Social licence’, ‘social capital’ and ‘social contract’ variously describe acting fairly and responsibly, as well as in compliance with laws; acting “ethically”; and maintaining trust of relevant stakeholders, be they customers, consumers or citizens. However, there is little reason to believe that organisations will be able to unravel high level statements of these concepts, or principles developing these concepts, into concrete frameworks, governance forums, processes and methodologies for making fair and ethically based decisions within those organisations. Organisations need guidance and tools to aid them. Guidance and useful tools are now emerging. However, disproportionate effort is being placed into creating what is now a plethora of statements of ethical principles for data driven services, algorithmic applications and AI. Not nearly enough effort is going into ‘making it real’, regardless of whether the ‘it’ is framed as digital trust, ethics, fairness, social licence, social capital or human rights. We need to refocus upon developing practical tools and methodologies for use in our workplaces today to ensure that we appropriately assess outputs and outcomes from computationally inferred behaviours of humans using diverse data, algorithmic decision making and other applications of artificial intelligence.

Keywords: social licence, social capital, social contract, corporate social responsibility, data ethics, algorithmic fairness, algorithmic discrimination, digital rights, human rights, AI ethics, digital ethics, digital trust