Today, major tech corporations are deploying computing clusters with millions of powerful GPUs to train AI models. However, in Africa, a continent with 54 countries and over 1.5 billion people, scientists have access to fewer than a thousand GPUs to build local language databases. UN Envoy on Technology Amandeep Singh Gill attributed this to the concentration of technological power and wealth in a few regions of the US. This is reported by Ixbt.com news.
At a conference at Seattle University, Gill called 2026 a turning point for AI regulation. According to him, technologies are moving past simple evaluation stages to become fully autonomous systems that perform complex tasks without human intervention. For example, Anthropic's Mythos model was released with limitations due to cybersecurity concerns. The diplomat warned that in the future, AI could become a source of systemic risk, bypassing cyber defenses and destroying social trust through deepfakes.
Another serious problem is the negative impact of data centers on the global climate. The energy required to run large language models and autonomous AI agents threatens carbon neutrality programs. The amount of fresh water used to cool servers, the disposal of obsolete equipment, and the costs of mining rare minerals are constantly increasing. Most regrettably, this environmental burden falls primarily on low-income developing nations.
To rectify the situation, a special scientific group of 40 experts has been formed under the UN, led by Nobel laureate Maria Ressa and renowned scientist Yoshua Bengio. To prevent the dominance of any single nation, the US and China were each given two seats, while the remaining 36 seats are distributed among other countries, including representatives from Africa. The group's first comprehensive report is expected to be published in July 2026.
Read “Zamin” on Telegram!
