Sign In for Full Access

Quick access through the institutional single sign-on Manchester Met Sign In
Skip this for now
Public Access Here

Sign In for Free Access

Login with email for free guest access to a range of Rise content
Logging You In!
Incorrect Password (Click Here to Reset)! Passwords Must Match Password must be more than 8 characters
Skip this for now
Man Met Access Here


Emily Rand & LOTI / Better Images of AI / AI City / CC-BY 4.0 

In this sprint, you will explore the policy conversations happening globally around establishing ethical frameworks and standards for AI. 

The UK took an initial step towards AI governance by releasing a white paper outlining five ethical principles that existing regulators across industries can refer to when evaluating AI use cases. Rather than heavily regulating AI technologies, the UK government intends to take a flexible approach underpinned by five principles: 

Safety, Security and Robustness

This principle states AI systems must reliably function without harm throughout their lifecycle. Companies must proactively identify and mitigate risks via rigorous testing and audits. The goal is preventing unintended consequences while ensuring continual beneficial performance. 

Appropriate Transparency and Explainability

Here the focus is ensuring AI decision-making processes are sufficiently visible and understandable to relevant stakeholders, depending on the system’s level of risk and influence. Enough information should be provided to monitor for reliability and fairness issues. 


This covers preventing unfair bias, discrimination, or rights deprivation. Precise thresholds vary across AI application areas, but in general biased outcomes disproportionately harming groups or individuals are prohibited. 

Accountability and Governance

The emphasis here is on company responsibility and oversight to ensure ethical AI development and deployment end-to-end. Following standards around documentation, assessments, and maintaining clear accountability allows issues to be redressed and safeguards to kick in over time. 

Contestability and Redress

This principle sanctions regulatory and public scrutiny, challenges to potentially unfair, biased or dangerous AI systems. Efficient mechanisms for raising issues and investigating algorithms that cause harm should exist, allowing fixes, compensation, or appeal of decisions where warranted. 

You can read the full paper here: pro-innovation approach to AI regulation.  

Other world leaders are drafting their own AI regulations. Conduct some research on AI legislation in the EU, China, and USA. Questions to investigate: 

  • What obligations do the regulations place on AI providers and users? 
  • Which AI use cases do they deem high-risk? 
  • How do their governance approaches compare to the UK’s five principles?