London: Google with its Gemini and Open AI with its ChatGPT seemed to have hit a homerun with their AI chatbots. But recently the tools have come under fire for inaccurate, offensive, and at times racial and sexist responses to certain key queries.

The BBC have reported that the recent instances of Google’s Gemini giving overtly incorrect political responses or ChatGPT requiring human instances for bias responses have a big problem of Bias data.

According to the report, the problem these tech firms face is that AI tools are trained on enormous amounts of data available publicly. This data also includes bias opinions and viewpoints on a large scale.

Once these AI tools pick up on bias responses it becomes difficult to differentiate unless human interference arises. This has led to inaccuracies like Gemini giving images of Asian women as Nazi soldiers in World War II.

Google and Open AI have clamped down upon reducing bias responses but considering the complexities of human history the responses on such issues become absurd. While, Google is confident of solving the problem in weeks, the Silicon Valley thinks this a huge headache which can be an achilles heel in AI’s growth.