Artificial Intelligence and Machine Learning

Perspectives on AI from Around the Globe

orbit paths around the Earth, illustration

The societal impact of artificial intelligence (AI) is global, yet the narrative around AI often centers around perspectives from the U.S. and Europe. Unsurprisingly—in a tech ecosystem that spans the planet—AI is powering change everywhere. International researchers are making AI breakthroughs in sectors as varied as healthcare, fintech, education, and environmental science, and the potential of Large Language Models (LLMs) is being investigated from India to Tanzania. However, country-level dynamics—including socioeconomics and linguistic diversity—can shape a nation’s AI priorities, flashpoints, and aspirations.

Seeking a ‘snapshot’ of the global outlook for AI, we asked experts to share their thoughts about developments and challenges in their countries.


“New technologies and advances that have generated most impact around the world have also generated similar impact in Brazil,” says Fabio Gagliardi Cozman, director of the Center for Artificial Intelligence at the University of São Paulo. “Recently there has been huge interest in generative models such as GPT or DALL-E.”

Cozman reports a “flurry of activity” in AI innovation for three of Brazil’s largest sectors (finance, healthcare, and agribusiness), and flags governmental efforts to establish research centers and enhance funding for AI innovation. Initiatives also include the Latin America Artificial Intelligence Index, which is collecting data about AI readiness across the region, he says.

Brazil has a population of around 200 million and enormous natural resources; however, the country has struggled with inequalities and preserving its “natural wealth and biodiversity,” says Cozman. AI technologies that “better inform the populations, speed up decisions, customize educational and health support, and improve sustainable production” are ripe for future development, he concludes.


Generative AI is also part of a “big conversation” in India, according to Urvashi Aneja, founding director of the Goa-based Digital Futures Lab. In particular, she says, focus is on “the use of large language models and generative AI tools for bridging both the digital and the linguistic divide.” However, Aneja points out that India’s diversity makes LLMs a “very complex problem.” According to the annual reference publication Ethnologue, there are 424 living indigenous languages in India. “Languages are not digitalized to the same extent, and users on the Internet do not represent the diversity of the Indian languages,” says Aneja.

India’s AI aspirations reflect universal trends; however, Aneja highlights several country-specific use cases, including using AI to improve high-risk labor practices. “Manual scavenging is still something that is done in the country. How can we automate some of those dangerous professions where there is no dignity of labor and a huge threat to safety and well-being?”

India’s size and “level of digital penetration” make the country well-poised to reap economic benefits from AI, Aneja says. “There is a very strong narrative in India that we missed the boat on previous industrial revolutions, and we cannot afford to miss the boat on this one.” However, this is somewhat troubling to Aneja: “We’re putting a lot of money and emphasis into training use cases of AI, but not enough into building guardrails, and institutional and regulatory capacity to manage harms and risks.”

Aneja also expresses concern there is “very little public oversight or transparency” on engagement between tech companies and the state. However, she suggests a lack of digitalized data in some Indian contexts could prove advantageous, prompting the creation of data in a “participatory, bottom-up way.” The country has a strong existing network of civic and grassroots organizations that can be tapped into for data creation. “We have an opportunity to build more problem-specific, purpose-driven, curated, representative datasets,” she says.


Chika Yinka-Banjo, a professor of computer science at Nigeria’s University of Lagos, reports that Natural Language Processing (NLP) and LLMs are currently the “most common” AI research fields in Nigeria. “These technologies can be applied to various sectors, including customer service, content creation, and language translation, which could be beneficial in the diverse linguistic landscape of West Africa,” she says. However, like Aneja, Yinka-Banjo raises a concern: “Developing AI applications that can effectively serve diverse communities, languages, and cultural contexts is a challenge that needs to be addressed.”

Financial technology (fintech) is a huge sector in Nigeria. According to a report by African tech startup portal Disrupt Africa, the country has “the most-populated fintech space on the continent,” with growth being driven by startups and investment—in the last 8.5 years [to July 2023], 41% of funding secured by African fintech ventures “went into Nigeria-based companies.” This makes fintech a very significant sector for AI innovation, according to Yinka-Banjo, along with education technologies, agriculture, and healthcare. “In Nigeria and West Africa, AI is being used for medical diagnosis, personalized treatment plans, and to improve access to medical services, especially in remote areas,” she says.

AI faces similar challenges in Nigeria as it does in other parts of the world, yet certain issues have a local texture. Says Yinka-Banjo, “In some cases, accessing high-quality, representative data may be a challenge, particularly in regions where data infrastructure is still developing, like ours. Lack of these data hinder the growth of AI in our region.” She highlights collaboration, investment, and fostering a responsible regulatory environment as key to the future of AI in Nigeria and West Africa.


Neema Mduma, a computer science and machine learning expert at the Nelson Mandela African Institution of Science and Technology (NM-AIST) in Tanzania, explained that that country’s government is prioritizing tech investment and actively integrating AI into its platforms. The judiciary, for example, is incorporating AI into its new transcription and translation system, which “will undergo training using Swahili dialects from diverse communities, spanning from Tanzania mainland to Zanzibar, as well as Tanzanian English attributions,” Mduma says.

AI solutions also are being implemented in education and agriculture—Mduma’s own research currently centers on developing AI tools to help smallholder farmers better control pests and diseases.

She also flags AI for rural healthcare. “Tanzania is experiencing a shortage of qualified healthcare professionals, particularly in the rural areas,” she said, highlighting opportunities for AI to provide preliminary diagnoses and reduce burdens on doctors.

Local flashpoints and geopolitics

Concerns around privacy, data bias, and cybersecurity reverberate globally. Sanmay Das, a professor of computer science at George Mason University and chair of the ACM Special Interest Group on Artificial Intelligence (SIGAI), also points to regulation and geopolitical competition as AI challenges for the U.S, along with potential disruptions in the hardware supply chain, “This could have serious ramifications in the industry,” he says.

There is significant discussion of AI legislation in Brazil, according to Cozman, although he points to a “more challenging” flashpoint in the country: “Recent elections have been affected by human and automated fake news, and it seems that AI-generated misinformation may be getting more sophisticated.”

In a U.S presidential election year, this also resonates for Das. “I think that the issue is really scale and magnification, rather than it [AI] being new; it might get attention in a specific context where it becomes a political football. That could lead to issues as well.”

On a connected planet, there are clear universal trends in the AI outlook, yet some national distinctions—driven by factors like geography, cultural diversity, and socioeconomics—do appear. However, the starkest divergences may only emerge as countries choose whether to take a collaborative or competitive approach to AI regulation and guardrails, a complexity that is just starting to unfold.

Karen Emslie is a location-independent freelance journalist and essayist.

Join the Discussion (0)

Become a Member or Sign In to Post a Comment

The Latest from CACM

Shape the Future of Computing

ACM encourages its members to take a direct hand in shaping the future of the association. There are more ways than ever to get involved.

Get Involved

Communications of the ACM (CACM) is now a fully Open Access publication.

By opening CACM to the world, we hope to increase engagement among the broader computer science community and encourage non-members to discover the rich resources ACM has to offer.

Learn More