ALLAI makes responsible artificial intelligence the norm
Artificial intelligence (AI) is developing quickly. Technology is improving all the time, yielding smarter algorithms and devices. As a result, AI is playing an increasing role in our daily lives, often without us being aware of it. Personalised suggestions on Netflix, automated processing of mortgage applications and chatbots that answer customer enquiries on line: all are examples of AI in action. However, as well as bringing convenience, efficiency and sustainability, AI raises questions and introduces challenges. Is everything ethical? And what are the risks? As co-founder of the Alliance for AI (ALLAI), Catelijne Muller provides high-level advice on the use of Responsible AI. Catelijne talks about ALLAI's mission, the importance of responsible AI and the support provided by SIDN Fund.
ALLAI is a vehicle for bringing together the business community, policy-makers, social partners, NGOs, public bodies and citizens to make AI more responsible. "We want people to be aware that a lot of decision-making is nowadays fully automated," says Catelijne. "And we accept things from algorithms that we'd never accept from a person." Catelijne therefore believes that a balanced debate is needed on the subject of responsible AI. "The current debate often focuses on extremes. You hear people saying things like 'AI is taking over' or 'AI can't yet do anything.' Or 'Europe is way behind the US and China in the AI race.' But AI isn't a race! The issue isn't who is developing most quickly, or whether Europe can keep up with China and America. What we should be asking is how we can work together on responsible AI."
So how can we work together on responsible AI?
As members of the EU Higher Level Expert Group on Artificial Intelligence, Catelijne and her partners Aimee Wynsberghe and Virginia Dignum advise the European Commission on the use and development of responsible AI. The fifty-two-member multinational Expert Group recently published the EU Ethics Guidelines for Trustworthy AI. "Stakeholders recognise the social importance of Responsible AI and welcome the publication of such guidance. However, they aren't really sure how to put it into practice."
Catelijne and her partners have therefore developed the ALLAI Responsible AI Program, which is designed to help businesses and other organisations devise responsible AI strategies. The Program provides a practical framework for translating ethical principles into action. It consists of three modules. The first is mainly about building awareness within the organisation, through collective examination of the technical, legal, ethical and social opportunities and challenges. The theme of the second module is the assessment of AI within the organisation. An organisation-wide survey is carried out to build a map of AI applications. That serves as the basis for assessing the ethical and legal status of those applications, and their impact, both on the organisation's own staff and more widely. The third module covers the development of a responsible AI strategy. Advice and tools are provided for defining and implementing internal procedures.
"Promotion of the Program has been confined to social media so far," explains Catelijne. "But we're already getting responses and enquiries from businesses and other organisations all around Europe, including Germany, Italy, Sweden and Norway. We're talking to numerous interested parties about how we can align the Program with their requirements. Not all of them are ready for all three modules yet, but some are."
SIDN Fund is supporting the ALLAI initiative. Which is not surprising, given that the Fund's annual theme for 2018 was Responsible AI. "Artificial intelligence is a very topical issue," says the Fund's Programme Manager Mieke van Heesewijk. "And rightly so, given the extensive use of AI and its social impact. A lot of AI isn't visible, but hidden within applications. So it's very important that AI is used in a responsible and ethical way. ALLAI is in dialogue not only with technical people, but also with people from a variety of other backgrounds: entrepreneurs, administrators, politicians, scientists and even artists. Together, these are the people who will shape the world of the future, where AI is extensively deployed. In SIDN Fund's view, Catelijne, Aimee and Virginia can play a key role in facilitating the multi-stakeholder debate that's needed in that context."
Catelijne and her team have gratefully used the grant from SIDN Fund to get their organisation off the ground and put it on the national and international map. So far, ALLAI has been launched at the WorldSummit AI in October 2018, and a strategy has bene developed and presented on various platforms. In addition, a first event has been organised, a website built, an Advisory Panel set up and the first Board and Advisory Panel meetings held.
"Last year, our aim was to identify a number of organisations that were ready to commit to Responsible AI and work with us on realisation," recalls Catelijne. "Although we ultimately want to be a global network, interest in the ALLAI Responsible AI Program has already taken us well past our first milestone. We're thrilled that so many businesses and other organisations are approaching us and saying that the Program is just what they're looking for. High-level talks are very important, of course, but changing things on the ground is ultimately what it's all about. That's where the effect of AI is felt."
Does your organisation use smart algorithms and big data? Want to know more about the responsible development and application of AI? Visit www.allai.nl.