The non-profit World Benchmarking Alliance (WBA) announced today the creation of the first multi-stakeholder coalition on ethical artificial intelligence (AI). The Collective Impact Coalition already has 39 supporting members, which include lead investors Fidelity International, Boston Common Asset Management, and 29 financial institutions representing USD 6.3 trillion of assets under management.
This coalition aims to raise awareness of the importance of responsible and ethical AI, increase understanding of the state of play and leading best practices, and improve technology companies’ commitment to ethical AI.
While the benefits and potential of AI are recurrently publicised by digital companies, more work needs to be done in identifying and disclosing its risks, especially around privacy and human rights. This issue was flagged by the WBA in its latest Digital Inclusion Benchmark, published in December 2021, after assessing and ranking 150 of the world’s most influential technology companies based on their commitments to advance a more inclusive digital society. Out of the 150 companies, to date only 20 have disclosed their commitment to follow ethical AI principles.
‘We believe that collective action is the way for digital companies to publicly commit to follow ethical principles for AI. If companies are not held to account, the consequences can be devastating for the users and society overall, and also lead companies to face reputational and revenue losses” said Lourdes O. Montenegro, Digital Sector Transformation Lead at the World Benchmarking Alliance.
Christine Brueschke, Sustainable Analyst and Digital Ethics Lead at Fidelity International, said: “As stewards of capital, we are acutely aware of our responsibility to our clients and wider society. How we hold investee companies to account today will help shape a more sustainable tomorrow, and we take this seriously. The growth of artificial intelligence will be exponential in the upcoming years and although it brings many opportunities for investors and society overall, it also carries considerable risks.
“These risks cannot be overlooked, and it is our belief that digital companies should be at the forefront of developing AI technology that is responsible and ethical. By working with the WBA and the wider Collective Impact Coalition (CIC), we look forward to collaborating to bring about better acknowledgment and mitigation of risks associated with AI.”
Lauren Compere, Managing Director at Boston Common Asset Management, said: “As investors, it is important for us collectively to better understand the landscape of AI, and promote robust transparency and disclosure of companies’ involvement in AI and how they mitigate the risk while embracing the opportunities it provides. This starts with having a public ethical or responsible AI policy. We are delighted to be part of this important collective initiative.”
In April 2022, the coalition published a public investor statement calling for digital economy companies to demonstrate their commitment to abide by ethical AI principles in both development and application of this technology.
In August 2022, the WBA started to communicate directly with the 130 companies ranked in the Digital Inclusion Benchmark that hadn’t yet disclosed their efforts to follow ethical AI principles and received immediate responses from seven of those companies.
Over the coming months, the coalition’s financial institutions will actively engage with relevant technology companies in their respective portfolios to highlight that the lack of transparency and commitment to ethical AI is a key concern for investors.
Lourdes O. Montenegro added: “The World Benchmarking Alliance will continue to expand this multi-stakeholder coalition to those civil society organisations and relevant groups keen to help drive progress on ethical AI and digital inclusion. The main goal of this initiative is to increase companies’ accountability; we hope to be able to track their progress and to increase the number of digital companies making public commitments to ethical AI. This will be reflected in next year’s publication of the Digital Inclusion Benchmark.”