You can apply “hype cycles” to loads of things, “AI for good” included. In my experience I have found many use cases of AI for social impact to be rather trite. After all, existing AI broken down into its constituent parts is basically a dataset, a parametrisation, a loss function, and an optimiser. Hyperbole around AI also leads to the obfuscation of the risks, grand ethical debates, and the attribution of responsibility for poor products and services to be abstracted away under the moniker of it was the “AI”.
SwissCognitive Guest Blogger: Gregg Barrett, Chief Executive Officer, Cirrus – “A meaningful use of Artificial Intelligence – In Africa – AI Trajectory 2023+”
It brings to mind the Financial Modelers’ Manifesto. The manifesto was a proposal calling for greater fiscal and risk management responsibilities in the wake of the US housing market collapse and the subsequent financial crisis of 2008-2009. It was created by two financial engineers Paul Wilmott and Emanuel Derman. Both Paul and Emanuel understood that models are not reality. A model describes the likeness of something. To describe the likeness of something is an approximation. It is not that something exactly. AI as it exists today is a model of something. Models contain an error – that needs to be accounted for.
That said over the last decade our capacity to model vast amounts of data with better function approximators has improved significantly. It is these improvements in machine learning that we now term AI. Critically, the proper application of these capabilities has the potential to save lives and impact society in positive ways that are not artefacts of marketing spin.
Societal impact in an African context
Governance and governing institutions in Africa are generally not of the same calibre as those found in the Western world, and law enforcement institutions do not have the same level of capabilities, specifically technological capabilities. With limited capabilities of law enforcement undesirable activities can and do proliferate. All too often societies in Africa are heavily reliant on the work of NGO’s to fill the void. Yet these NGO’s themselves are no panacea as the requisite technological platforms are in most instances beyond their reach impeding the effectiveness of their operations.
As an example, a leading anti-human trafficking organisation operating at one of South Africa’s busiest airports, while intercepting trafficked persons uncovered systemic corruption and organised crime involved in the trafficking networks. Their airport security clearance was revoked preventing them from operating in the airport. Their cost per intercept is now around 300 USD which needs to be reduced to around 50 USD to be in-line with funding – requiring the adoption of AI methods which can be scaled without a resulting scaling in cost and headcount. While realising the need for AI in this context is easy, bringing it to pass is anything but. Indeed, the challenges of the undertaking are more than an excuse to do nothing, but to do nothing is costing lives – an unacceptable situation given that it is solvable
Working with data management and AI platform providers there is now an undertaking to place anti-human trafficking organisations in Africa onto the same technological footing as their Western counterparts. In parallel are efforts to secure anonymous mobility data from the major telcos to help in identifying choke points for placing new stations with human monitors to intercept trafficked persons. Following the implementation of the requisite technology platform, the strategy is to start work with the South African banking sector on a Financial Intelligence Unit. The purpose of this unit will be to train financial services and anti-money laundering staff on a new platform that allows these professionals to share knowledge, information, and best practices in real-time. Further, this unit is intended to help survivors of human trafficking get access to banking services that they would not otherwise qualify for because of poor credit and other issues related to their trafficking experience. Relatedly, there is work to significantly bolster the data science / machine learning of the anti-human trafficking organisations by establishing collaborations with academia and industry, and the formation of a data hub to provide key data to researchers, academics, law enforcement officers and others seeking to deepen knowledge and understanding in the fight against human trafficking.
Thank you for reading this post, don't forget to subscribe to our AI NAVIGATOR!
Beyond the anti-human trafficking the intention is to extend the data management and AI platform to other areas, including: water and environmental resource management; wildlife trafficking and environmental crimes; arms trafficking and conflict finance; exploitation of natural resources (illegal mining, logging and deforestation, illegal and unregulated fishing). An undertaking to produce synthetic rhino horn that is contaminated to cause issues like indigestion when consumed is also in the early stages of formation. The data management and AI platform applicable to anti-human trafficking are needed to provide insights into the illicit supply networks into which the synthetic contaminated product will be introduced.
Menstrual health and related areas have not been effectively addressed in developing markets, and there is a need to unite the academic / research community, medical practitioners, NGO’s and enterprises with women in the developing world to deliver effective outcomes.
The groundwork is now being laid for the development of a women’s health application under a public benefit organisation that is freely available to all women. Importantly, the undertaking is architected from the start to enable data generation and where the data generated is properly governed and used to support research and guide interventions.
Data management and AI platforms now provide us with the capabilities to better address data and intelligence asymmetries where diverse collaborative efforts can be united under a single application. Some high-level targets include: development of a cost-effective test for endometriosis; cost-effective hormonal cycle tracking; provision of sanitary vending in corporate workplaces; encouraging the use of and provision of menstrual cups and applicators; provision of expert help to those struggling with menstrual health and related issues (at no cost) like PMS, PMDD, fertility etc.
It is only through cross-sector collaborations bringing together even broader coalitions to facilitate analysis of data, that we can have a bigger impact on menstrual health. Helping to inform stakeholders of who needs help and what help they need, resulting in more effective resources allocation. This contrasts with blindly giving funding to stakeholders who in turn purchase more resources. The importance of these collaborations cannot be overstated when addressing social issues, rather than a landscape in which organizations each carve out their own territory.
The implementation of the social impact strategy requires the establishment of a philanthropic engineering operation funded through revenues from commercial engagement with industry. The philanthropic engineering operation is intimately connected to the commercial operation, and develops deep, hands-on, and often long-term relationships with NGOs and social sector organisations. This includes regularly sending engineering teams to work with these organisations and collaboration on developing solutions. The on-site engineering helps to develop foundational understanding of the respective fields enabling deeper relationships and the establishment of broader collaborations of players to work on solving social problems.
In the context of anti-human trafficking and menstrual health in Africa, there is the capability to drive on-the-ground action with partners that have the capacity to use utilise data management and AI platforms for maximum impact, with organisations that have: sources of data; will benefit from machine learning analysis of the data; have staff in place to act on the insights to drive action.
While the application of AI through a philanthropic engineering operation will directly improve and save human lives, the engagements will also have a powerful impact internally, helping to attract, retain, and engage employees. The challenges being tackled require world class engineers and in Africa there is a dire need to create opportunities to cultivate this talent – to offer meaningful work and the chance to make a difference. Simply, there is no shortage of people to train, and no shortage of opportunities for them to make an impact.
No multi-continental corporation has the capacity to impact anti-human trafficking or menstrual health when acting alone. However, when leaders in industry, academia, and NGO’s share data to address the problem collaboratively, combined with AI, real progress can be made.
Notably the anti-human trafficking effort now underway was inspired by an incredible woman who sadly passed away in 2020 – and her sister is the catalyst for the menstrual health work. It serves as a reminder that while technology enables new capabilities, it is only out of the vision of a heart for others that lives are transformed. Without the leadership of these two incredible women the anti-human trafficking and menstrual health work in Africa would not be where it is today, and the application of AI to this work would not be possible.
A reminder that heart to transform lives is something no AI model will ever replicate.
- AI and the North – South divide
- AI as a technology ecosystem
- AI benchmarks and environmental impacts
About the Author:
Gregg Barrett is a seasoned executive with extensive and diverse experience in strategy, building and managing relationships, deal-making, communication, developing high-performance teams, organisational leadership, and problem-solving across a range of areas. Over the last decade, Gregg has led work in data science, machine learning, corporate research, and corporate venture capital. This includes the establishment and management of data science, machine learning, corporate research, and corporate venture capital operations, working across people, processes, and technology, integrating structured and unstructured data to direct research, business, and investment strategy.