Use of AI could worsen racism and sexism in Australia, human rights commissioner warns.

In order to be as accurate and efficient as possible, artificial intelligence systems should be trained using content from any field, any culture, and, above all, free from discriminatory bias. Developing an AI tool without paying attention to the type of material it is fed means obtaining a flawed product, subject to the opinion of the average human, incapable of broadening its horizons beyond the knowledge it already possesses.

According to Lorraine Finlay, Australia's Human Rights Commissioner, the use of unregulated artificial intelligence could dangerously exacerbate racism and sexism within the nation. A study dating back to May 2025, for example, has already highlighted how candidates interviewed by AI recruiters are often discriminated against by the tool in question in cases of different accents or disabilities.

Furthermore, in order to obtain the most suitable results for the Australian population, especially in medical contexts, it would be wiser to use artificial intelligence trained on data from the country, without relying on tools tailored to US citizens.