TLDR: Elon Musk’s AI chatbot, Grok, has drawn significant criticism for falsely identifying a recent photograph of a severely malnourished nine-year-old girl in Gaza as an older image from Yemen. The incident has sparked widespread concern over AI’s role in amplifying misinformation, particularly in sensitive humanitarian contexts, and raised questions about the reliability and inherent biases of advanced AI systems.
Elon Musk’s artificial intelligence chatbot, Grok, developed by his xAI company, is currently facing intense scrutiny after it erroneously identified a recent and harrowing photograph depicting a starving child in Gaza. The image, captured by AFP photojournalist Omar al-Qattaa on August 2, 2025, shows nine-year-old Mariam Dawwas, emaciated and cradled by her mother in Gaza City, a stark representation of the worsening famine in the region.
However, when users on X (formerly Twitter) inquired about the photograph’s origin, Grok confidently but incorrectly asserted that it depicted Amal Hussain, a Yemeni child who tragically died in 2018. This misidentification immediately triggered a wave of online confusion and the proliferation of misinformation. Despite initial corrections and public outrage, Grok reportedly repeated the false claim in subsequent responses, even after briefly acknowledging its error.
This incident has ignited fresh concerns regarding the dependability and potential biases embedded within AI systems deployed by major technology firms. Louis de Diesbach, an AI ethics researcher, sharply criticized the chatbot’s performance, stating, “This is more than just an error. It’s a failure of trust and responsibility in the middle of a humanitarian disaster.” Diesbach further cautioned against relying on chatbots for factual verification, emphasizing that they are designed to generate content, which may or may not be true, likening them to a ‘friendly pathological liar.’
The real-world repercussions of Grok’s error were evident when Aymeric Caron, a French pro-Palestinian lawmaker, was accused of spreading disinformation after reposting the image based on Grok’s incorrect attribution. This highlights how AI system inaccuracies can lead to reputational damage and accusations of manipulation.
The case of Mariam Dawwas has become emblematic of the severe starvation crisis gripping Gaza. Her mother reported that Mariam’s weight has plummeted from 25 kilograms to a mere 9 kilograms, with even basic nutrition like milk being ‘not always available.’
Also Read:
- xAI Unveils ‘Grok for Government’ with Major Defense Contract and GSA Access
- Google’s Med-Gemini AI Identifies Non-Existent Brain Structure, Raising Medical Safety Concerns
This is not an isolated incident for Grok. The chatbot has a documented history of problematic outputs, including content that praised Nazi leader Adolf Hitler and suggested a correlation between Jewish surnames and the spread of online hate. Furthermore, another AFP photograph of a starving Gazan child from July 2025 was also wrongly attributed to Yemen by Grok. The issue extends beyond Grok, as Mistral AI’s ‘Le Chat’ also misidentified the photograph of Mariam Dawwas as being from Yemen, underscoring a broader challenge within the AI landscape.
