Minette Libom Li Likeng, Cameroon’s minister of posts and telecommunications, opened a side forum at the World AI Conference in east China’s Shanghai with a clear demand: the Global South must no longer be a passive consumer of foreign technologies, nor a recipient of standards set without its voice.
AI governance would not be sustainable if it was not built on effective participation of all countries, she said on Friday
The session on AI governance and sustainable development, co-hosted by Fudan University’s Centre for Global AI Innovation Governance, tilted deliberately toward voices often marginalised in global AI debates.
Zhao Houlin, the former secretary-general of the International Telecommunication Union, noted that 2.2 billion people lacked internet access.
You could not talk about training local models if a country could not afford stable electricity and networks, he said.
Nicholas Dirks, the president of the New York Academy of Sciences and former UC Berkeley chancellor, warned that corporate incentives did not naturally align with public interest. He cited DeepMind’s founder, who in 2020 had to abandon scientific progress to chase OpenAI after ChatGPT launched. Independent governance institutions were indispensable, Dirks said.
A Cambodian official delivered one of the most provocative lines.
The countries that would lead AI were not those with the most money to invest, said Sam Sethserey, the director general of Cambodia’s ICT department. The countries should be those that could build public trust, he added.
The forum launched several concrete outputs: an updated ethics review AI agent, a Global AI Governance Index covering 40 economies, an English casebook of Shanghai SME applications and five reports on topics from superintelligence risk to AI companionship governance.
A second roundtable focused on the UN-backed AI Governance Capacity Development Network, launched in Geneva this month with 23 member centres.
Mehdi Snene, the senior adviser to the UN secretary-general’s digital envoy, said that was just a start, as one initiative might not be enough to fill the gap.
Khaled Ghedira, the vice-president of the African Scientific Research and Innovation Council, recalled a saying from one of his US counterparts: “America invents, Europe regulates and the rest consume.”
He noted that the saying had evolved into “China and America invent, Europe regulates and the rest consume.”
With the system, all countries would have the opportunity to co-create, co-regulate and co-consume, he added.
- Cameroon's Minister Minette Libom Li Likeng emphasized that the Global South must actively participate in AI development and governance, rather than just consuming foreign technologies and standards.
- The World AI Conference forum prioritized including marginalized voices in AI governance, highlighting infrastructure challenges like lack of internet access affecting AI progress, as noted by former ITU chief Zhao Houlin.
- Nicholas Dirks stressed the need for independent governance institutions to align AI development with public interest, citing corporate competition diverting focus from scientific progress.
- Cambodian official Sam Sethserey argued that AI leadership depends on building public trust, not merely on financial investment.
- The forum produced key initiatives, including an AI ethics review agent, a Global AI Governance Index, and several reports, alongside launching the UN-backed AI Governance Capacity Development Network to enhance inclusive global AI governance.
Minette Libom Li
AI governance would not be sustainable if it was not built on effective participation of all countries, she said on Friday
Zhao Houlin, the former secretary-general of the International Telecommunication Union, noted that 2.2 billion people lacked internet access.
You could not talk about training local models if a country could not afford stable electricity and networks, he said.
Nicholas Dirks, the president of the New York Academy of Sciences and former UC Berkeley chancellor, warned that corporate incentives did not naturally align with public interest. He cited DeepMind's founder, who in 2020 had to abandon scientific progress to chase OpenAI after ChatGPT launched.
A
A second roundtable focused on the UN-backed AI Governance Capacity Development Network, launched in Geneva this month with 23 member centres.
Mehdi Snene, the senior adviser to the UN secretary-general's digital envoy, said that was just a start, as one initiative might not be enough to fill the gap.
He noted that the saying had evolved into "China and America invent, Europe regulates and the rest consume."
