Algorithms: Public sector urged to be open about role in decision-making
Public sector bodies must be more open about their use of algorithms in making decisions, ministers have been told.
A government advisory body said greater transparency and accountability was needed in all walks of life over the use of computer-based models in policy.
Officials must understand algorithms’ limits and risks of bias, the Centre for Data Ethics and Innovation said.
Boris Johnson blamed a “mutant” algorithm for the chaos over school grades in England this summer.
Ofqual and other exam regulators across the UK were forced to back down following a public outcry over the use of a computer program to determine A-level and GCSE grades after the cancellation of exams.
The regulator’s chief executive resigned after the algorithm used to “moderate” marks submitted by schools and grading centres resulted in nearly 40% of them being downgraded, in some cases by more than one grade.
It was accused of breaching of anti-discrimination legislation and failing to uphold standards.
The government was forced into another U-turn last month over aspects of its planning reforms after Tory MPs accused ministers of relying on a faulty computer-based formula to decide house building targets across England.
In a new study, the Centre for Data Ethics and Innovation said there needed to be greater awareness of the risks of using algorithms in make potentially life-changing decisions and more done to mitigate them.
Those running organisations, it said, had to remember they were ultimately accountable for all their decisions, whether there were made by humans or artificial intelligence.
Its recommendations include requiring all bodies to record where algorithms fit into their overall decision-making process and what steps are taken to ensure those affected are treated fairly.
While organisations should be actively collecting and using data to identify bias in decision-making, it said there was a risk techniques used to mitigate bias, such as positive discrimination, could fall foul of equality legislation.
It urged the government to issue guidance on how decision by algorithm must comply with the Equality Act.
Adrian Weller said there was an opportunity for the UK to demonstrate global leadership in the responsible use of data and ensure appropriate regulatory standards were in place.
“It is vital that we work hard now to get this right as adoption of algorithmic decision-making increases,” he said.
“Government, regulators and industry need to work together with interdisciplinary experts, stakeholders and the public to ensure that algorithms are used to promote fairness, not undermine it.”
The Information Commissioner’s Office urged organisations to consult guidance on the use of artificial intelligence.
“Data protection law requires fair and transparent uses of data in algorithms, gives people rights in relation to automated decision-making, and demands that the outcome from the use of algorithms does not result in unfair or discriminatory impacts,” it said.