The increasing sophistication of artificial intelligence (AI) has raised complex issues around the responsible use of the technology. A recent case that has drawn considerable attention involves Grammy Award-winning singer Taylor Swift, who found herself the target of AI-generated explicit images. These images illustrate the formidable capabilities of AI when misused by cybercriminals.

AI-generated explicit images of pop star Taylor Swift swept across social media platform X, illuminating the disturbing potential of AI technology. Suspected to be created using Microsoft Designer, the fabricated images not only caused a furore among Swift's fans and privacy advocates but also underscored the growing challenge posed by AI-generated fake pornography.

According to Michal Salát, Avast Threat Intelligence Director, the newfound ease of generating such realistic images using AI mirrors the impact Photoshop had years ago on image manipulation. "This is just a different photo editing software, in a way," highlighted Salát. The primary difference now lies in the accessibility provided by AI technologies, prodding society to adjust its ways of regulating these tools.

Of particular concern is the ethical dilemma posed by the use of AI to generate explicit images, especially when the images feature identifiable individuals without their consent. Salát notes, "There’s probably nothing inherently wrong or ethically problematic with AI-generated porn. But the ethical problem, at least for me, is that you can relatively easily generate an image with a known or a specific face on it."

This unnerving potentiality is not just limited to celebrities such as Taylor Swift; essentially, anybody could be targeted. Despite the illegality of "revenge porn" in New Zealand, legislation covering AI-generated explicit content remains absent, underlining the urgent need for law-making to keep pace with advancing technologies.

Arguably, AI companies could introduce safeguards against the creation of unconsented explicit content. Salát suggests simple measures could include prohibiting the generation of images featuring specific individuals or using customer-submitted images as source materials. While some users might find ways to bypass these controls, adding such measures would make the misuse of AI much more difficult.

Salát emphasises, "The way I see it, at the moment, it’s virtually impossible to avoid this — you can only make it harder. I think the point I’m trying to make here is the companies that offer these services should try harder to avoid this.” There remains, however, the chance of an individual creating a personal model without any restrictions in place, though Salát cites that this requires substantial technical knowledge and computing power.

The security expert describes the current state of AI development as comparable to the evolution of the security industry, terming it a "start over with AI". The takeaway from the disturbing incidence involving Taylor Swift is the pressing need for responsible AI usage. Balancing the potential of AI with ethical considerations and security measures will be crucial as the technology advances. The perspectives of experts like Michal Salát can guide us towards a more informed and cautious approach in navigating this intricate field.