In the rapidly evolving landscape of artificial intelligence, a recent roundtable hosted by Chamber and UKAI brought together policymakers, tech experts, and advocates to unpack the critical challenges facing AI development.
AI bias remains a real risk for the growing industry, to ensure AI can be used positively to promote economic growth and improve services, AI experts argue that the root problems of bias and inequality must be addressed.
The roundtable, chaired by former shadow secretary of state for culture, media and sport, Thangam Debbonaire, discussed the challenges and opportunities AI represents, the government’s approach to AI bias, and the ethical implications of AI bias and political communication.
Debbonaire was joined by Samantha Niblett MP, Dan Aldridge MP and Chamber and UKAI founder Ben Howlett who shared their expertise with the panel to answer the question: How do we ensure AI is used to help develop a fairer, more efficient society, rather than harm it?
AI is a tool: ‘It’s exciting, but not at the expense of people’
The speakers discussed the role AI can play in building the UK’s future, with the technology driving innovation in healthcare and supporting NHS services as part of Labour’s Plan for Change as well as supporting green industries and building a skilled labour force.
However, the panel emphasised that the development and implementation of AI cannot be left unchecked, and that efforts must be made to ensure processes that involve AI remain equitable.
Cecilia Jastrzembska, Director and President of the Young European Movement, spoke of the risks of AI bias in hiring and automatic filtering systems, discussing claims of instances of AI bias and discrimination against women applicants on job postings based on their names, downgrading their applications based on the AI’s assumption of gender.
She explored AI bias further through a case-study of policing in America where she claims AI that was used in a predictive policing scheme in black and ethnic communities, where it occasionally identified false risks and “self-perpetuated” its own mistake, embedding societal inequalities into its policing recommendations.
One particularly illuminating example from the healthcare sector highlighted the complexity of AI bias. Junior doctors enthusiastically implemented an AI diagnostic tool which they cited as greatly improving their care and efficiency, while senior doctors remained resistant, citing the tool as “rubbish”. This created a philosophical dilemma that underscored the challenges of AI integration, many people simply do not see or understand the crucial role AI will play in the future.
To overcome this the panel agree that the root-causes of AI bias and inequality must be addressed, with attention paid to ensure gender parity and wider diversity in the AI space, with Jastrzembska saying it is about “who trains the data and what [they train]”.
Combating AI bias and diversity head on
Addressing AI bias requires a comprehensive strategy that goes beyond surface-level interventions. The roundtable participants emphasized the critical importance of examining and diversifying the entire AI development ecosystem.
This means focusing intensely on the input data – recognizing that AI systems are fundamentally a reflection of their training. The experts advocated for “gender parity in policy setting and in inputting training data”, questioning not just what data is used, but who is responsible for collecting and curating it.
The goal is to create a more inclusive approach to AI development, ensuring that training datasets represent diverse perspectives and experiences. This involves bringing more women, people of color, and individuals from varied backgrounds into AI development teams, policy-making roles, and decision-making positions.
Moreover, there’s a recognition that bias mitigation is not a one-time fix, but an ongoing process of critical evaluation. As one participant noted, companies will typically “do as much as they can get away with”, underscoring the need for clear, robust regulatory frameworks that establish explicit boundaries and accountability mechanisms.
The ultimate aim is to create AI systems that don’t just avoid perpetuating existing biases, but actively work to counteract them, promoting fairness, equity, and genuine technological innovation.
‘We have a role as politicians’: Communicating ethical and practical use of AI
AI continues to face a “communication challenge” in Britain, Aldridge mentions that in many instances “the public are fearful” of AI, and that when politicians “knock on the door and talk about tech [the public] don’t say anything positive about it. It’s how social media’s ruined our lives. It’s not a tool of empowerment”
As he noted, the public isn’t pressing for huge conversations about AI – they’re more concerned with immediate issues like fixing potholes, getting GP appointments, and addressing local problems.
The government has a crucial role in bridging this gap between AI tech leaders and the public. Speaking on this challenge, Dobonnaire said: “I think we have a role as politicians to be saying to voters, to the public, to businesses here’s a load of good stuff for AI can do, and one of things that AI can do is free up your MP to do other stuff that’s even more important to them. That’s a good thing frankly.”
However, Aldridge also explains that as an MP, you become a “jack of all trades, master of none”, with little time to delve deep into new tech developments and AI tools, even if they could greatly benefit their work. He provides some advice for those seeking to engage with policymakers: don’t just offer a product demo. Instead, align communication with MPs’ core concerns, be persistent, and demonstrate how AI can solve practical problems for them and their constituents.
Final Thoughts
As the UK positions itself at the forefront of AI development, the diversity roundtable highlighted both the challenges and opportunities ahead. The path forward requires a collaborative approach – bringing together technologists, policymakers, and diverse voices to ensure AI serves the broader public interest.
The message was clear: AI is not something to be feared, but a powerful tool to be carefully and thoughtfully developed. With the right approach, it has the potential to solve complex problems, drive economic growth, and create meaningful change.
