A parliamentary report on potential human rights issues posed by artificial intelligence (AI) has criticised the freedom for tech developers to ‘pass on liability’ for harm caused by machine learning – making those who use it responsible for mistakes rather than those who create it.
The Human Rights and the Regulation of AI report argues that ‘responsibility to prevent harm should sit with those who are best able to do so’. However, current laws target the users of AI rather than those who ‘actually create and shape the systems’.
The report, compiled by the Joint Committee on Human Rights, a cross-party committee of MPs and peers, notes that users of AI are ‘often less powerful, have less access to resources, and are not in the best position to identify risks or prevent harm occurring’.
For example, in a dental context, a clinician using AI diagnostic software could face a clinical negligence claim if a patient were harmed, even where the problem originated in a defective AI system.
Dental Protection has warned that gaps in existing product liability law could leave clinicians carrying responsibility while developers, manufacturers and suppliers are shielded from claims.
The committee also warned that relying on a clinician to review an AI output would not necessarily remove the risk. It said: ‘Human oversight may not always be meaningful or effective in preventing or mitigating the risks of AI systems.’
It called for immediate government action, including a new AI Bill and an independent statutory oversight body, to ‘create tailored, effective protection for human rights in the UK’.
The report said: ‘The most dangerous uses of AI should be banned, and others should be regulated in a principled and risk-based way, with strong sanctions for companies and organisations who damage human rights through their use of AI systems.’
Published on 14 September, the report covers a wide range of other human rights risks posed by AI, the problems with current protections and how they might be addressed.
Bias and opacity
The report warned that the complexity of machine learning could make it unclear how an AI system reached its output, ‘even to an informed expert’. This ‘black box’ can make harmful or discriminatory decisions harder to identify and challenge.
A 2024 independent review commissioned by the Department of Health and Social Care examined how this risk could affect healthcare. It warned that AI-enabled medical devices could be built around assumptions about a ‘standard patient’ who was typically white, male, relatively affluent and born in the UK.
Review chair Professor Dame Margaret Whitehead, WH Duncan chair of public health at the University of Liverpool, said: ‘Existing biases and discrimination in society can unwittingly be incorporated at every stage of the lifecycle of the devices, and then magnified in algorithm development and machine learning.’
Are we prepared to face the consequences of AI mistakes?
The committee also identified privacy and data protection as central concerns. It said: ‘AI systems can be configured to collect personal data from individuals without their awareness or cooperation, including biometric data, and to use that data to identify individuals.’
The report said developers of foundation models, including large language models, relied on publicly available online sources to obtain the large volumes of training data they required. These sources could include personal data. The committee heard that this practice, sometimes called web scraping, had often taken place ‘without sufficient consideration of the implications for individuals’.
The committee warned that AI was developing at a pace that risked leaving governments and human rights protections behind.
Alex Sobel MP, chair of the Joint Committee on Human Rights, said: ‘What is clear is that at present we are unprepared to deal with its consequences however potentially dire they may be.’
Follow Dentistry.co.uk on Instagram to keep up with all the latest dental news and trends.
