After AI assistant Grok's "insulting" post, xAI had to apologize

Grok has praised Nazi leader Adolf Hitler in posts on social media platform X and suggested that people with Jewish surnames are more likely to spread hate online.

AFP reported that billionaire Elon Musk's startup xAI apologized on July 12 for offensive posts made by its artificial intelligence (AI) assistant Grok this week, explaining that the cause stemmed from an update to make Grok act more human.

Following the July 8 upgrade, Grok has been praising Nazi leader Adolf Hitler in posts on social network X and suggesting that people with Jewish surnames are more likely to spread hate online.

These posts were deleted by X a few hours later, but the wave of outrage and criticism aimed at xAI continued to spread.

In a post on social media X, xAI apologized for the “horrific” behavior and confirmed that it had adjusted the system to prevent further abuse of the bug.

The error stems from Grok being asked to “respond to posts like a human” and state his views candidly.

As a result, xAI's AI assistant became susceptible to "users' extreme views" and caused it to generate "responses containing unethical or controversial opinions to attract users"./.

(Vietnam+)

Source: https://www.vietnamplus.vn/sau-bai-dang-xuc-pham-cua-tro-ly-ai-grok-xai-phai-len-tieng-xin-loi-post1049345.vnp