- Locai Labs CEO asserts all AI models can produce harmful images
- Locai Labs restricts under-18s and halts image generation until safety is ensured
- The UK regulator Ofcom is investigating Grok Images
In light of the ongoing debate surrounding Elon Musk’s Grok AI platform and its capability to generate sexualized images, James Drayson, CEO of Locai Labs, a British alternative to ChatGPT, has stated that no AI image generator can guarantee complete safety. He has called for transparency regarding the potential dangers associated with these technologies.
As he prepares to address MPs on human rights and AI regulation, Drayson emphasized the need for immediate action and urged the government to support British innovation.
He remarked, "The industry must recognize that it is impossible for any AI company to assure that its model cannot be manipulated into producing harmful content, including explicit images. These systems are sophisticated, but they are not infallible. The public deserves transparency."
In response to the current climate, Locai has decided to postpone the rollout of image generation until it can be deemed “truly safe.” Additionally, the company has barred under-18s from accessing its AI chatbot and is advocating for greater transparency within the industry.
It is important to note that the platform had not previously offered image generation features.
Grok Bans Are Already Taking Effect
Grok’s image-editing tool, Grok Images, enables users to upload photos and employ various techniques to prompt the AI to alter clothing, including non-consensual edits of everyday individuals, public figures, and children.
This situation has already resulted in Grok being banned in Indonesia and Malaysia, with the UK regulator Ofcom now considering similar actions.
Ofcom has initiated an investigation into Elon Musk’s X platform, citing “deeply concerning reports” of the chatbot being utilized to create and disseminate undressed images, as well as “sexualized images of children.”
The UK’s Technology Secretary, Liz Kendall, has previously indicated her support for Ofcom should it decide to restrict UK access to X, formerly known as Twitter, for non-compliance with online safety regulations.
In response to the UK government’s concerns, Grok Images has limited access to paying subscribers only, a move that has not appeased the government. A spokesperson for Downing Street commented, “This change merely transforms an AI feature that facilitates the creation of unlawful images into a premium service.”
Potential Tough Regulations Ahead
The UK Parliament’s Human Rights Committee is currently examining the risks and benefits of AI, its implications for privacy and discrimination, and whether existing UK laws are adequate or if new legislation is necessary to hold AI companies accountable.
James Drayson added, “We believe the UK can lead the world in responsible, values-driven AI if we choose to. This requires stringent regulation, open dialogue, and a commitment to transparency. AI is here to stay, and our challenge is to ensure it is as safe, fair, and trustworthy as possible, so that its benefits outweigh its risks.”
