Digital Ethics Summit 2022
Last week, techUK hosted our sixth annual Digital Ethics Summit, hosting speakers and attendees at our offices and across the world on an online stream. The day was full of vibrant discussions, bringing together the perspectives of policymakers, tech experts, regulators, academics and civil society representatives. Below you can find a short summary of the day’s sessions and recordings of every session will be available early in the new year!
This year’s Summit conversations reached far and wide, beginning with a keynote by Paul Scully MP, the UK Minister for Tech and the Digital Economy, who emphasised the need for citizen trust to further innovation. One way to do this will be through AI assurance, and the Minister launched the latest report from the Centre for Data Ethics and Innovation on the topic, sharing a year’s worth of research on the Barriers and Enablers to AI Assurance. He was followed by Dr Janil Puthucheary, the Singaporean Senior Minister for Communication and Innovation, who shared details of the fascinating initiative developed with industry called AI Verify; the world’s first AI governance testing framework to demonstrate responsible use.
We then moved on to the day’s panels, kicking off by asking the question whether tech regulation is raising the bar for digital ethics. The panel discussed the extent to which the proliferation of regulation is encouraging organisations to focus on ethics, or whether compliance takes precedence and ethics is put to one side. Panellists explored the extent to which regulation can help us address ethical conundrums, and also how clear regulation may help companies understand and reach minimum ethical benchmarks – and in turn, the public identify responsible tech practices – in turn increase public trust by making it easier to identify the companies doing so. However, it was also argued that complying with regulation will not be enough to make the tech industry ethical, as a lot of ethical judgment also goes in to interpreting laws.
Following the morning plenary, we moved into the first round of breakout sessions, one of which focused on the latest developments within online safety legislation. Here, a packed room listened to our panel of researchers and legal experts, who shared an overview of the international context, discussed the progress of the UK's landmark Online Safety law, unresolved policy and compliance concerns in the UK, as well as practical takeaways based on extensive academic research and industry experience.
In the other morning breakout session we explored whether progress is being made on digital ethics in the public sector. The panel agreed that we are seeing progress, especially on the understanding of data ethics, however on broader digital ethics there was still some way to go. The panel also raised questions about how to increase accountability to ensure public bodies follow ethical practices, and emphasised that we also need to ensure the public gets to benefit from the opportunities offered by the use of technology.
In the second plenary of the day, we looked into how the tech industry can learn from mistakes made in the past. Panellists questioned who gets to decide what counts as a mistake, and shared what they saw as a lack of incentives to share lessons from failures externally. It was also highlighted that sometimes industry clearly does learn and improve, for example in reducing algorithmic biases, but that this usually gains little awareness compared to the original issue, which reduces interest in sharing such learning. Panellists also highlighted the role of regulation in furthering transparency in how industry is making progress following what could be described as ethical failures.
We were then joined by two consecutive keynote speakers. First, Elham Tabassi, Chief of Staff at ITL, NIST, for a keynote on the role of standards in increasing trust in AI, shared the latest on international AI standards development. Then, Professor Luciano Floridi argued that ‘the new challenge is not technological innovation, but the governance of the digital’ and outlined his recent work on ethics-based auditing of AI to address this challenge.
In the third plenary of the day, we focused on how breakthrough technologies can be deployed to benefit humanity. A central point to the discussion was that an optimistic approach is needed to get the balance right between advancing technology and benefits to humanity. From digital twins to help climate scientists to data to combat financial fraud and develop life-saving medicines, the panel agreed that ethical risks must be challenged head-on to maintain public trust in the process.
In the second round of breakouts, one session focused on the responsibilities of the tech industry in respecting human rights, and how human rights relate to digital ethics. The panel explored how ethics can be embedded in the development of products from the ideation stage through the use of human rights impact assessments. They also discussed the complexity of responding to rapidly evolving social and political contexts, as well as understand how to identify and mitigate risks through the entire lifecycle of tech products.
The other afternoon breakout session looked into the emerging metaverse worlds, and what we can do to keep people safe there. The panel discussed the inequity of different societal groups in the metaverse and the discrimination that certain groups in society face, and reflected on the extent to which forthcoming online safety legislation will help to address this. The panel also touched on the effectiveness of AI in moderating content, and how to differentiate what is accessible to different age groups.
In the final panel of the day, we focused on the question of AI ‘sentience’ – and specifically to what extent we are drawn to the sentience narrative over other pertinent AI questions. One view represented was that sentience is not relevant and discussing it feeds unhelpful hype, but it was also argued that people are in fact rather easily convinced of sentience – as witnessed by the number of people engaging with online bots. The panel was however overall in agreement that the pace of improvement of tools such as ChatGPT means this discussion will become increasingly pertinent, and safeguards need to be put in place to help people verify the authenticity of content sources.
At the very end of this year’s Summit, we were brought to a close by techUK’s deputy CEO, Antony Walker, who alongside Professor Luciano Floridi reflected on the themes of the day. Professor Floridi also announced a new research project to be launched in January 2023, where a research group will be developing a new ethical framework that can be used to draft future EU legislation for the metaverse. More details will be available early next year, which techUK will make sure to share with our networks.
The techUK team would like to thank everyone who joined this year’s Summit, especially our fantastic speakers, and we look forward to welcoming you all again next year!
If you are interested in getting involved, please email [email protected] and [email protected].
Margot Stumm
Margot joined techUK in September 2022 as the Head of Events and Sponsorship, running the event programme for our flagship events, such as the techUK Annual Dinner, Defence Ethics Summit, Building the Smarter State etc
