AI is adopting humanity’s worst habits:UN report suggests that artificial intelligence is learning about sexism, racism, and objectifying women

Artificial intelligence has quickly become part of daily life. People now use tools like ChatGPT and other AI assistants to write emails, create presentations, plan marketing campaigns, and even solve everyday problems. But as AI becomes more common, experts are raising a serious concern: it is also learning and repeating human biases. Ahead of the United Nations’ Global Dialogue on AI Governance and the AI for Good Global Summit in Geneva next month, UN Women has urged governments, tech companies, and AI developers to make gender equality a key part of how AI is designed, trained, and regulated. In the UK alone, nearly 88% of advertising and media agencies already use AI in some form, making the issue more urgent than ever. AI still sees women through old stereotypes In some cases, AI has even generated responses that portray women as sexual objects or as inferior to men. Researchers also found that when AI was asked to complete simple sentences mentioning a person’s gender, around one in five responses contained sexist or misogynistic language. Some responses even described women as property rather than individuals. This isn’t a technical glitch UN Women says these problems are not random mistakes. Instead, AI is producing biased results because it has been trained on decades of human-written content that already reflects unequal treatment of men and women. AI is making online abuse worse The risks go far beyond unfair stereotypes. Women and girls already face high levels of online harassment, and AI is making some forms of abuse easier to create and spread. UN Women found that nearly one in four women human rights defenders, activists, and journalists surveyed had experienced AI-assisted online violence. Around 12% said their personal photos were shared without permission, while 6% reported being targeted through AI-generated deepfake images or videos. As AI-created content becomes harder to distinguish from real content, experts fear that harassment, fake content, and image-based abuse will become even more difficult to stop. Too few women are building AI Another major concern is that women remain underrepresented in the AI industry itself. According to the International Labour Organization, women make up only 30% of the global AI workforce. UN Women warns that when the people creating AI systems do not reflect the diversity of society, the technology is more likely to overlook or ignore the needs and experiences of many users. Without greater participation from women and other underrepresented groups, today’s biases could become permanent features of tomorrow’s AI. Women could face bigger job losses Fair AI is good for business too Building fair and inclusive AI is not only about protecting rights it also makes business sense. Research from the UN Women’s Unstereotype Alliance found that advertisements free from gender stereotypes perform better. Brands using inclusive advertising saw stronger sales growth, higher customer loyalty, and better pricing power than competitors. To help companies reduce bias, the Alliance introduced a new playbook in June 2026 that helps marketers identify and remove bias before AI-generated content is published. The future of AI depends on the choices we make UN Women believes AI has enormous potential when developed responsibly. It can help detect stereotypes instead of spreading them, improve accessibility, and ensure broader representation. However, these benefits will only become reality if women and girls are included at every stage from designing and training AI models to deploying and governing them. As global leaders prepare to meet in Geneva, UN Women’s message is simple: if women are left out of shaping AI today, the inequalities of the past could become deeply embedded in the technologies of the future.