Elon Musk pushes back against Grok AI image abuse claims, ‘It will refuse to produce anything illegal’

2 hours ago 2

ARTICLE AD BOX

Elon Musk has dismissed allegations that Grok produces illegal images of minors, saying the AI only responds to user prompts and refuses unlawful requests, while acknowledging that rare misuse attempts are treated as bugs and fixed promptly.

Elon Musk has rejected claims that his artificial intelligence system Grok generates illegal images of minors, as several countries move to restrict access to the tool over concerns about obscene content.(AFP)

Musk denies allegations

Posting on X on Wednesday, Musk said he was “not aware of any naked underage images generated by Grok” and insisted there were “literally zero”. He argued that Grok only creates images in response to user prompts and is designed to refuse any request that violates the law.

Musk added that while attempts to exploit the system through adversarial prompts may occasionally produce unexpected results, such issues are treated as bugs and addressed immediately.

Indonesia and Malaysia impose restrictions

Despite Musk’s assurances, Indonesia and Malaysia have become the first countries to restrict access to Grok. Indonesia’s Ministry of Communications and Digital Affairs announced a temporary ban last Saturday, citing concerns about the risk of fake pornographic content produced using AI technology.

The ministry said the move was intended to protect women, children and the wider community. It has also asked X to provide immediate clarification on the issue.

Communications Minister Meutya Hafid said the government viewed non-consensual deepfake sexual content as a serious violation of human rights, personal dignity and national security in the digital space.

Malaysia followed with similar action on Sunday. The Malaysian Communications and Multimedia Commission said it was limiting access to Grok until effective safeguards are put in place. Notices were issued to X Corp. and xAI LLC, but regulators said the responses failed to address the inherent risks posed by the AI tool.

Image generation feature curtailed

The restrictions come after xAI, which owns Grok, decided last week to limit its image-generation capabilities on X. The move followed widespread criticism after the tool was found to generate obscene images of women and children.

Under the new rules, only paid subscribers can generate and edit images. Previously, the feature was available to all users for free, subject to daily limits.

Regulators say changes fall short

Several regulators, including in the UK, have criticised the paid subscription requirement as inadequate. They argue that the standalone Grok app, which operates independently of X, still allows users to generate images without subscribing, leaving gaps in oversight.

Key Takeaways

Elon Musk claims Grok AI is designed to refuse illegal image generation.
Countries like Indonesia and Malaysia are taking proactive measures to restrict Grok due to concerns about AI-generated obscene content.
The debate over AI regulation highlights the need for effective safeguards in rapidly evolving technologies.

Read Entire Article