FEATURED
CURRICULUM

Stanford course will critique AI through lens of ‘African intellectual sovereignty’

Share to:
More options
Email Reddit Telegram

Africa map and AI on a smartphone; Lalocracio from Getty Images Signature/Lalocracio from Getty Images.Canva Pro

Course is built on assumption that such ‘unjust practices and injustices’ exist, scholar says

Students in a spring 2027 Stanford University course will critique AI systems through the lens of “African intellectual sovereignty” and “unlearn … epistemic injustices.”

The course will examine how to gain “African intellectual sovereignty in an era dominated by algorithmic governance, data colonialism, and techno-capitalism,” according to the course description.

Students will “work to unlearn and deconstruct epistemic injustices, dismantling unjust practices, assumptions, and institutions while actively creating new ways of knowing and being.”

Further, the description states that the course critiques “Eurocentric epistemologies embedded in AI systems and global knowledge infrastructures” and elevates “Indigenous African knowledge systems.”

The class is offered by Stanford’s Department of African and African American Studies and taught by Professor Samuel Nkansah. 

The College Fix reached out to Nkansah for more information on the course, but did not receive a response. 

The Fix also asked Stanford University if the course requires students to adopt specific political or ideological positions. The school also did not respond.

However, National Association of Scholars Senior Fellow Lilla Kiss raised several concerns about this course in an interview with The Fix. 

NAS is a non-profit that advocates for intellectual freedom, virtuous citizenship, liberal arts education, and reform in higher education.

“Stanford’s description starts with labels such as ‘data colonialism’ and ‘techno-capitalism’ without clearly explaining what would justify those labels or call them into question,” Kiss said. 

Further, “Having less access to technology does not automatically mean a country is being ‘colonized,’” she said. 

What’s more, many African countries are not lagging technologically. 

She cited the United Nation’s 2024 E-Government Development Index, which examines digital progress trends in member countries. South Africa, Mauritius, Tunisia, Morocco, Seychelles, and Egypt all have “E-Government Development Index (EGDI) values above the world average,” meaning they are not technologically lagging. 

She also said the course seems to wrongfully assume the cultural or technological homogeneity of Africa. 

“The course description also indicates that Africa is a ‘homogenous’ place, with shared victimhood and problems. It is not. Indeed, African countries differ considerably,” she said. 

“Stanford’s framing therefore risks overlooking differences among African countries and how foreign technology can actually help communities build something of their own,” she said. 

For example, Meta’s open source Llama “gives developers a foundation they can modify for local needs without financing the original model’s development.”

Kiss also noted that the course description signals a clear “ideological direction before an open classroom discussion even begins.”  

This raises concerns over whether students will be able to push back against the course’s assumptions or be expected to adopt its conclusions, she said. 

Kiss also raised concerns that “the course is built on the assumption that such ‘unjust practices and injustices’ are existing without examining if this presumption stands at all.”

Dictating to students what they should reject without fairly considering competing arguments “risks turning education into ideological instruction,” she said. 

“Finally,” Kiss said, “African and European intellectual traditions deserve equally rigorous scrutiny, including an examination of who gets to define ‘African values’ and how the course accommodates Africans who might dare to disagree.”

Stanford University is currently embroiled in a controversy after it admitted to using AI to alter students’ appearances in promotional materials, including digitally transforming one male student into a black female, The College Fix reported.

University spokesperson Charlene Gage told Inside Higher Ed that the school violated its own AI policy and has removed the banners with the AI-edited images. 

MORE: New Jersey university trains counseling students to be ‘social justice advocates’