How AI bias is impacting the Global South in new wave of 'techno-colonialism'

The lack of inclusiveness and representation in the development of artificial intelligence is growing and leading to increasingly biased technology solutions against mainly non-western and developing countries, according to the United Nations.

Carme Artiga Brugal, co-chair of the AI High Level Advisory Body of the UN, says that the technology output we have today is being created in the interest of the Global North, which generally includes developed countries in North America and Western Europe but also Japan, Singapore, Australia and New Zealand.

The result is that countries in Latin America, Africa, the Middle East, and greater Asia are adopting global AI policies that might work against them.

"This exacerbating of biases and lack of inclusiveness of data leads to missed opportunities for these countries," Ms Brugal told The National on Thursday at Abu Dhabi Finance Week.

This can include blocking jobs prospects for regional talent, the adoption of global AI policies that work against the development of local economies, and environmental solutions generated by data that serves some countries and harms others.

"We need an inclusive approach, that is good for equity, for effectiveness and efficiency ... that have curated data, representative data, and also talent from local countries that find the solutions to the problems they have," said the co-chair of the AI High Level Advisory Body, composed of 32 experts from government, private, and civil society sectors who make and advance recommendations for international governance of AI.

Ms Brugal said this exclusion of the Global South from global decisions on AI "is a new way of techno-colonialism" during her talk at ADFW.

"I developed the solution, I developed the technology, and I think that the rest of the world are going to be the consumers of those solutions," she said about the current state of global AI policy outlook during her discussion titled Setting International Rules on the Use of AI.

A select few countries are leading the roughly 40 international AI conferences and initiatives around the world, according to the UN's AI High Level Advisory Body.

"We discovered it was only seven countries in the world, the G7 [Canada, France, Germany, Italy, Japan, the UK and US] are party of all of them, but there are 118 countries in the world that are party of none among those – especially in the Global South," she said.

The UAE, which ranked fifth on the Global Vibrancy Tool 2024, participated in only two of these global initiatives, said Ms Brugal. "So the UAE being the leader of the Global South [in AI] ... it's not even sitting in the table for the most relevant discussions around AI policy."

Data control

The advancement of AI is predominantly driven by the private sector. Tech giants including Google, Microsoft and OpenAI lead in global AI research, development, and therefore its deployment, according to the World Economic Forum.

While transparency is key to the accountability of AI policies, it relies on data accessibility which tends to fall in the hands of a few.

This was highlighted by Bhavish Aggarwal, the chief executive and founder of India's largest mobility platform, Ola, in a recent interview.

He said that while India is a top producer of the world's digital data, which is estimated at 20 per cent – matching its approximate make up of the world population – it only stores one tenth of that amount.

"Ninety per cent of India's data is exported to global data centres, which are predominantly owned by big tech companies," he told Reuters news partner, Asian News International, in a July interview. The exported amount is then processed by these companies, and then sold back to Indian institutions in dollars, he added.

"Novo-colonialism. Techno-colonialism. Techno-colonialism. Yeah. I don't know whether this is a word or not, but ... it's been done."