President Macron protesting his own retirement reform?
Your eyes do not deceive you.
Amidst the ongoing troubles in France over the retirement reforms and the social unrest that has decried from President Macron’s controversial bill, Internet users have turned to AI to get creative.
Some impressively realistic images generated by artificial intelligence tools have been circulating on social networks for the past few days.
They come courtesy of the latest version of Midjourney, an artificial intelligence program created by a San Francisco-based independent research lab, Midjourney Inc, which first went live in March 2022. The program generates images from descriptions called “prompts”, in the same way Stable Diffusion and DALL-E create images.
The latest version (the fifth) was made available on 16 March in open beta ie: open to anyone who is interested in using the product, and the evolution is first-rate. And since all it takes is one sentence to create any image, Internet users have naturally jumped on the opportunity to keep it topical and let off some steam by prompting images reflecting the current context in France.
That's how you can across images on social networks of Macron sitting on piles of rubbish, as a bin man or demonstrating against his own pension reform.
Check them out for yourself.
Users have also kept it topical in the US, with images circulating online of Donald Trump being arrested. Ever since the former president warned that he expected to be indicted this week, in the case involving a bribe paid with campaign funds in 2016 to porn star Stormy Daniels, users have been dreaming of what could be...
Thanks to AI, dreams really do come true…
The results are really quite impressive, and mildly worrying.
Indeed, despite the inherent humour that decries from these absurd images which obviously cannot purport to be reality, the advances the AI tools are making in leaps and bounds once again stress that if these deepfakes look this convincing, the technology can be used to spread disinformation or commit fraud.
Only recently, several reports have emerged of the US Federal Trade Commission warning consumers to beware of the rise of AI-powered tools like ChatGPT, Microsoft's Vall-E and specifically the voice-cloning app by ElevenLabs' AI speech software, VoiceLab. This technology can convincingly clone voices to dupe victims into believing their loved ones are in trouble and need cash fast.
At this point, AI dreams become the stuff of nightmares.