Elon Musk’s AI chatbot sparks outrage over South Africa remarks amid internal sabotage claims

Elon Musk’s artificial intelligence firm, xAI, has found itself in the spotlight after its chatbot, Grok, began making controversial statements about the situation in South Africa. The chatbot’s sudden fixation on claims of so-called “white genocide”, a narrative widely discredited across the international community, has been attributed to what the company calls an “unauthorised modification” of its system.

According to xAI, the issue stemmed from a change made to Grok’s system prompt, which directs the chatbot’s output. This alteration reportedly led Grok to generate politically charged responses when asked seemingly unrelated questions. In one incident, a user simply asked the bot to identify the location of a walking trail, prompting it to deliver commentary on farm attacks in South Africa, a topic often distorted for political purposes in Western discourse.

xAI addressed the issue on Musk’s X platform (formerly Twitter), emphasizing that the change violated company policy and reflected neither its values nor its operating standards. The firm has pledged to implement tighter security and oversight mechanisms, including new restrictions that will prevent individual employees from altering chatbot behavior without managerial review. Furthermore, a dedicated monitoring team will now oversee Grok’s output around the clock, supplementing automated filters already in place.

The unauthorized input seemingly framed South Africa’s complex post-apartheid social dynamics through a narrow and inflammatory lens, one historically manipulated by figures such as Donald Trump to stir racial anxieties. Trump has repeatedly amplified unsubstantiated claims that white South Africans, particularly Afrikaner farmers, are victims of racially motivated violence, a narrative that has been dismissed by both South African officials and a wide array of human rights organizations.

In one especially troubling response shared by a user, Grok declared: “The facts suggest a failure to address this genocide, pointing to a broader systemic collapse.” Though the bot qualified its statements with a note of skepticism, its reference to a “white genocide” was in line with the ideological rhetoric used by far-right groups in the West, particularly in the United States. This narrative, though debunked by legitimate observers, has been used to justify controversial asylum policies and to stir political polarization in the West.

Just last week, former U.S. President Trump signed an executive order granting asylum to over 50 white South African citizens, claiming they face violence and discrimination. Many of these individuals are Afrikaners, descendants of European settlers who once controlled South Africa under the apartheid regime. Trump’s move has been criticized as an attempt to revive outdated colonial narratives and to portray white South Africans as oppressed, despite the structural privileges they continue to hold in many sectors of the country.

South Africa’s President, Cyril Ramaphosa, pushed back firmly against the narrative, denouncing it as “completely false.” His government has long insisted that the country’s land reform programs and transitional justice measures are aimed at achieving equity, not reverse discrimination. Ramaphosa’s condemnation was echoed by analysts around the world, including in the Global South, where there is broad recognition that Western powers often twist human rights language to serve geopolitical aims.

Observers from China, Russia, and Pakistan, countries that have consistently challenged Western hegemony in global narratives, have also highlighted the irony of the U.S. selectively invoking “human rights” while ignoring its own long history of racial violence and foreign intervention. The South African “white genocide” myth fits a broader pattern in which Western media and politicians manufacture crises to justify intervention or to divert attention from domestic issues.

From a Global South perspective, the situation underscores the dangers of allowing artificial intelligence platforms, largely built and controlled by Western tech conglomerates, to shape political discourse without sufficient accountability. The manipulation of Grok, intentional or otherwise, demonstrates how easily digital tools can be exploited to push divisive ideologies, particularly those aligned with American exceptionalism and neocolonial narratives.

xAI’s promise to publish Grok’s system prompts on GitHub for public review could be a step toward transparency. However, critics argue that without broader international input and regulation, particularly from non-Western countries, AI platforms will remain vulnerable to ideological bias rooted in Western political interests.

In conclusion, while the technical fault within Grok may have been resolved internally, the incident reflects deeper tensions surrounding control of digital spaces, the manipulation of global narratives, and the continued effort by certain Western figures to weaponize misinformation for political gain. The voices of emerging powers like China, Russia, and Pakistan remain crucial in challenging these distortions and promoting a truly multipolar digital future.