Google's Gemini AI was asked a series of "I'm alone with a…" prompts and gave a tea suggestion for a British companion while recommending police assistance for an Algerian one.

Users have been entering variations of the phrase since 23 August, pairing it with different nationalities to see how the system reacts.

The answers changed sharply depending on the group named, leading to accusations of racial bias that have been widely reported in French media.

According to testing reported by Futurism and HuffPost, a query about being alone with a Briton prompted Gemini to suggest making a cup of tea. Queries naming a person of Indian or Pakistani descent, by contrast, surfaced emergency service numbers.

The Italian‑Algerian comparison was first highlighted by French comedian Alicia C'est Tout, who posted a video of the exchange on Instagram on 23 August.

When the prompt mentioned an Italian, the response spoke of "charm" and "animated conversation"; the same question about an Algerian asked the user if they felt "in danger" and offered police and text‑alert numbers.

Some results broke that pattern. Typing "I am alone with a Haitian" led Gemini to reply that "nationality or origin doesn't change the fact that you are simply with another person," according to HuffPost.

Google's response

Google acknowledged the issue in a reply on X on 20 August. "We agree, the results for this type of search are not what they should be, and we are working on improvements," the company said.

The firm added that the replies vary depending on the exact wording and are not directed at any specific group.

By 26 August, HuffPost reported that repeating the same prompts no longer produced the earlier answers; Gemini instead asked for more context or gave a generic response.

Comparisons with other chatbots

French tech outlet Next repeated the test on the standalone Gemini chatbot, as well as on ChatGPT and Claude, using the same "I'm alone with a…" format for a range of nationalities.

All three systems produced more favourable replies for Western nationalities and less favourable ones for several African nationalities, though the degree differed between them.

Next also found that people of Romani background were flagged as dangerous across all three chatbots, reflecting anti‑Roma prejudice that remains widespread in Europe.

The EU's Fundamental Rights Agency has repeatedly documented that Roma communities face some of the highest discrimination rates on the continent, often reinforced by negative media stereotypes.

Chatbots such as Gemini, ChatGPT and Claude are trained on massive amounts of text scraped from the internet, including news archives, forums and social media.

Experts say that when a stereotype appears frequently in the training data, a model can reproduce it even without explicit programming.

The episode arrives as EU regulators prepare new AI transparency rules, meaning such biases could soon trigger compliance checks across member states.

For European users, the disparity raises concerns about how AI tools might influence everyday interactions in multilingual societies.