The AI for Good Global Summit 2024 took place on May 30-31 in Geneva, bringing together a gaggle of over 2,500 participants representing some 145 countries.
In her opening remarks, ITU Secretary-General Doreen Bogdan-Martin set the tone for the event by explaining the necessity for inclusivity in AI development.
She said, “In 2024, one-third of humanity stays offline, excluded from the AI revolution, and and not using a voice. This digital and technological divide is not any longer acceptable.”
The summit showcased examples of AI applications, corresponding to Bioniks, a Pakistani-led initiative designing reasonably priced artificial limbs, and Ultrasound AI, a US-based women-led effort improving prenatal care.
These contribute to an unlimited body of projects that actually showcase how AI can speed up disease diagnosis, help develop latest drugs, provide movement to those that lost it through injury disease, and far more.
AI For Good also dived into how AI might help attain the UN’s Sustainable Development Goals (SDG), which set out broad and far-reaching plans to grow and modernize less-developed nations while alleviating poverty, climate change, and other macro problems.
Melike Yetken Krilla, head of international organizations at Google, discussed several projects where Google data and AI are getting used to trace progress toward the SDGs, map it across the globe, and collaborate with the World Meteorological Organization (WMO) to create a flood hub for early warning systems.
AI can also be helping conservationists protect the environment, from the Amazon rainforest to Puffins off British coastlines and salmon in Nordic waterways.
AI’s potential for good – as per the Summit’s sentiment – is clearly substantial indeed.
But as ever, there’s one other half to the story.
AI’s push and pull
Rather than one-way traffic, AI tempts to each shatter and speed up digital divides.
For one, there is powerful evidence that AI entrenches currently existing divisions between more and fewer technologically advanced countries. Studies from MIT and the Data Provenance Initiative found that almost all datasets used to coach AI models are heavily Western-centric.
Languages and cultures from Asia, Africa, and South America remain primarily underrepresented in AI technology, leading to models failing to accurately reflect or serve these regions.
Moreover, AI technology is pricey and hard to develop, and a select few corporations and institutions undoubtedly hold the vast majority of the control.
Open-source AI projects provide a lifeline to corporations globally to develop lower-cost, sovereign AI but still require computing power and technical talent that continues to be in high demand worldwide.
AI model bias
Another tension on this push and pull is bias. When AI models are trained on biased data, they inherently adopt and amplify those biases.
This can result in severe consequences, particularly in healthcare, education, and law enforcement.
For instance, healthcare AI systems trained predominantly on Western data may misinterpret symptoms or behaviors in non-Western populations, resulting in misdiagnoses and ineffective treatments.
Researchers from leading tech corporations like Anthropic, Google, and DeepMind have acknowledged these limitations and are actively searching for solutions, corresponding to Anthropic’s “Constitutional AI.”
As Jack Clark, Anthropic’s policy chief, explained: “We’re trying to seek out a strategy to develop a structure that’s developed by an entire bunch of third parties, quite than by individuals who occur to work at a lab in San Francisco.”
Labor exploitation
Another risk to harnessing AI for good is cases of labor exploitation for data labelers and annotators, whose task is to sift through hundreds of pieces of information and tag different features for AI models to learn from.
The psychological toll on these staff is vast, especially when tasked with labeling disturbing or explicit content. This “ghost work” is crucial for the functioning of AI systems but is regularly missed in discussions about AI ethics and sustainability.
For example, former content moderators in Nairobi, Kenya, lodged petitions against Sama, a US-based data annotation services company contracted by OpenAI, alleging “exploitative conditions” and severe mental health issues resulting from their work.
There have been responses to those challenges, showing how AI’s threat to vulnerable populations can, with collective motion, be stamped out.
For example, projects like Nanjala Nyabola’s Kiswahili Digital Rights Project aim to counteract digital hegemony by translating key digital rights terms into Kiswahili, enhancing understanding amongst non-English speaking communities in East Africa.
Similarly, Te Hiku Media, a Māori non-profit, collaborated with researchers to coach a speech recognition model tailored for the Māori language, demonstrating the potential of grassroots efforts to make sure AI advantages everyone.
A balancing act
The push and pull of AI’s advantages and disadvantages can be tricky to balance within the forthcoming years.
Rather than representing a brand new paradigm of international development, AI is a continuation of many years of discourse investigating the impacts of technology on global societies. It’s each highly universal and highly localized.
Large-scale AI tools like ChatGPT can provide a ‘blanket’ of encyclopedic knowledge and skills that billions can access worldwide.
Meanwhile, smaller-scale projects like those described above show that, combined with human ingenuity, we will construct AI technology that serves local communities.
Over time, the important thing hope is that AI will grow to be concurrently cheaper and easier to access, empowering communities to make use of it as they like and, on their terms, with their rights. Of course, that would also include rejecting AI altogether.
AI – each the generative models created by tech giants and traditional models created by universities and researchers – can actually offer societal advantages.
There is way to be skeptical and hopeful about. Such was the promise of other technologies before AI, from the printing press to the combustion engine.
AI might extend more deeply into society than other technologies, nevertheless it stays under human control for now.
