Over the last decade, artificial intelligence has become increasingly more sophisticated and successful in carrying out tasks previously thought only humans could do. Its applications such as recommendation systems, advanced search algorithms, image processing and facial recognition, are rapidly transforming businesses and whole industries such as healthcare, agriculture, transportation, and many more. Since the 1960s, AI has continuously raised questions and problems that could not be answered with technical solutions; the consequences, regulation, societal change, and ethical dilemmas.
With the support of The Analytics Institute of Ireland and in collaboration with global law firm DLA Piper, this Wednesday 22nd January, Women in AI Ireland hosted the first “Ethically Driven AI Design” event in Dublin which took place in Tangent, Trinity’s Ideas workspace, with more than 70 attendees actively engaging to discuss various ethical dilemmas and AI’s impact on the society.
The interactive format of this event characteristic to humanities and social sciences created an enabling environment for ideas and different standpoints to thrive: keynote speech followed by moderated table discussions covering 6 major topics currently being discussed in the field of ethics in AI: Healthcare and AI ethics, Self-driving cars, AI in advertisement and recommendation systems, Future of work and HR management, Economic wealth distribution and access to financial services, and Privacy and regulation of facial recognition technology.
“AI is an enormous opportunity for STE(A)M to thrive. Now more than ever we need people of all backgrounds, especially women and minorities to enter the field of artificial intelligence.”
Current thinking in ethics in AI and law
Introduced by Gareth Stokes and Mark Rasdale, our keynote speaker Treena Dunlea-Peatross, Legal Director on DLA Piper Ireland’s Intellectual Property and Technology team, gave a high level overview of the current thinking in ethical considerations for the design of AI from a lawyer’s perspective, drawing on the views expressed by various regulators and practitioners across the globe.
With an impactful opening, Treena invited the audience to think critically. AI technology is new but the society has already encountered many of those problems throughout history, and especially women’s history.
“Think about robot rights and ownership. Not so long ago the society was deciding on women’s rights. The same questions about property, authorship and rights the society was asking about women.”
To tackle the societal problems rising with AI, Treena highlighted the importance of ensuring that there is diversity of thought when designing and developing AI technologies. That’s the new and necessary approach. She touched on aspects of the current law in Ireland, the UK and the US and how they could apply to AI technologies, highlighting ethical issues for consideration along the way such as maker’s accountability and the impact on the minorities.
“Without diversity of thought and ideas, we risk building in bias into AI”.
Treena explained the difference of having and not having a regulation such as GDPR, and how developing technology impacts the user without such regulation in place. She then introduced various initiatives being taken at the global and national level to establish ethical frameworks for the development of AI technologies.
The good news is that AI technology is making us think globally — currently there is a big global focus on long-term sustainability and fairness in the use of emerging technologies.
Ethics in AI workshop bringing minds together
Treena’s speech raised important questions and ethics dilemmas that engaged the audience in reflective thinking and reasoning which they were about to practice in a group setting. The ethics workshop consisted of table discussions with designated topics, a moderator skilled in design thinking provided by DLA Piper and up to 10 participants from the audience.
Together with DLA Piper moderators the audience engaged in dynamic discussions on ethical considerations and the impact of AI on our the society from different angles:
Healthcare and AI Ethics with Conor McEneaney discussed embracing a wholly AI diagnostic system. It was important for all participants that there was some human involvement throughout, and the extent of this involvement has been widely discussed. Some members of the table foresaw the change in the role of a physician as becoming more of a communicator and the liaison between the technology and the particular patient. Some members brought up and idea of a new profession, that of a data nurse — someone with responsibility for administering the AI tools to a patient and interpreting these results with a doctor being consulted in extreme or unusual circumstances.
Self-driving cars group discussed the trolley problem with Edel O’Kelly. The group recognised the public sensitivity towards the number of car accidents involving self-driving cars and human-driving cars, even though autonomous cars could save more lives on the road compared to human-driving cars.
Mark Rasdale discussed Advertisement, Recommendation systems and Awareness. Interestingly, this group happened to form a random expert group with data scientists, machine learning engineers, philosophers and lawyers. They discussed the ethics dilemma of recommendation systems and feed algorithms forming a so-called“filter bubble” of the information a user is exposed to that is limited to their own interests and the interest of their narrow and similar network.
Future of work and HR Management with moderator Alice Duffy discussed the automation of tasks and full human roles. The participants pointed out that even some of their own roles in cybersecurity and data did not exist until short time ago. Inputting data in an AI system based on our digital personality or profile can be an inaccurate reflection of work capabilities. The group’s perception was that AI can achieve a more objective and unbiased outcome than human assessment of employee performance, rewards and recognition, and company KPIs.
Eileen Johnston considered Economic Wealth Distribution and Access to Financial Services and emphasised the potential of concentrating AI technology to specific organisations with unlimited workforce thus widening the economic gap. The group discussed the credit scoring system of China assigning people a financial score based on their good behaviour and pointed out that Western society is looking at the economy and rising ethical dilemmas through Western lenses.
Eilis McDonald and Cez Bicki focused their group discussion on Facial Recognition Technology (FRT): Privacy and Regulation. The group generally felt more comfortable with authorities using FRT and biometric data collection rather than companies who would mainly use it for marketing purposes. There was a consensus that obtaining consent for using FRT is a big challenge and that such consent might not always be informed.
The event concluded with a readout session from each table where the audience with moderators’ support had the opportunity to express their viewpoints, ethical concerns and valuable ideas on the design of AI systems, as well as learn the viewpoints of others. Alessandra Sala encouraged women to think big and speak up, calling women to enter, explore and contribute to AI ecosystem in Ireland with Women in AI.
Women in AI is the first global community of women in Artificial Intelligence with over 3,000 members in 90 countries. We aim at shaping gender-inclusive AI that benefits global society by increasing female representation and participation in AI. Our global platform empowers female AI-visionaries and practitioners by providing access to educational resources, networking opportunities, and hands-on support, while accelerating a community of global pioneers in the field of Artificial Intelligence.
Article written by Iva Bubalo