3 days ago
EXCLUSIVE Aussie photographer Charles Brooks slams Elon Musk over billionaire's wild AI claim
An Aussie artist has lashed out after Elon Musk reshared artificially modified versions of his photos to 'promote' his chatbot.
Accomplished artist and musician Charles Brooks achieved international recognition in 2022 for his 'Architecture in Music' photographic series, capturing the interiors of musical instruments.
This week, he was shocked to learn from a friend that his work had been altered by a third party using artificial intelligence (AI) and shared by Musk on X – formerly known as Twitter.
Above his photos, edited to include little men inside his instruments, was Musk's invitation for users to 'Generate images with Grok'.
Mr Brooks insisted '98 per cent' of the images were actually generated by him.
'I went and did some digging, and eventually I found this thing he'd shared from a couple of days ago,' he told Daily Mail Australia.
Mr Brooks conceded that Musk was not 'quite' passing his photos off as Grok's work, but believes he was operating in a 'grey area'.
'The thing that annoys me is that my photos are being kind of used, and my name's not being attached to them,' he said.
Musk reshared three of Mr Brook's images which had been altered with AI, imploring X users to generate images with Grok
'And the way that they're being presented makes it look like they're AI photos.'
AI software can now produce essays, creative fictions, photos, videos and audio to boot. But the models have been 'trained' repetitively at first using real images, artworks, videos and soundbites.
The ethics of AI training and artistic intellectual property have been debated since the engines were first unveiled to the public.
Mr Brooks insists there was no mistaking the images were his.
'For me, the main thing is, these are real photographs of very specific instruments,' he explained.
'You know, there's a double bass in there that's photographed. And I know exactly what bass that is. That's a English Charles Therese Double Bass from 1860 that my friend plays in the New Zealand Symphony Orchestra. Beautiful bass.
'That's my photo of inside that specific instrument. And that should never be claimed as AI.'
Mr Brooks insisted he's not anti-AI and uses it for an array of tasks including managing his website and correcting writing.
He hopes the saga serves as a warning to artists – as the creative industries are forced to coexist with what could represent a disrupting force.
'If there's anything that I would wish out of this it would be that, you know, there's a way for works or images to always be tracked back to the artist when they're shared,' he said.
'AI is going to see this stuff, it's going to learn. So are people, for that matter, other people can look at my work and maybe figure out how I've done it and do something similar. Great, good on them.
'But my creations should be able to be tracked back to me easily and it's frustrating that it's so simple for people to lift them and share them, and you lose that chain back to the original artist.'
Mr Brooks was astonished about how little people seemed to care to credit real-world artists, whose time and effort became the foundation of every subsequent AI creation.
'I think with all of this new technology, if we can trace every bitcoin transaction on Earth, we should be able to trace a photo,' he said.
Many online were left shocked after Mr Brooks detailed the incident online.
'Elon didn't mention my name at all, Eric did put my name in but only as a comment which I find a bit disingenuous,' Mr Brooks claimed in a video posted on Reddit.
'I'm not terribly against people modifying the images as long as it's not for a commercial reason, but if you're going to do that, please credit me.'
The artist took his grievances to the platform after his X account was suspended. His account has since been reinstated.
Commenters unleashed against the technology billionaire.
'So sorry you have to do go through this crap with him. I've completely stopped using twitter, will never call it X,' one wrote.
Another added: 'I'm sorry Elon Musk is such a piece of sh*t. Your photography is really beautiful,' another added.
A third wrote: 'This is a salesman posting fake results about his product. So what does that say of the product?' another said.
Many viewers wondered if Mr Brooks had a legal case against Musk.
'I don't think (Mr Brooks) has a case against the 'original' Twitter poster, but (Musk) is using stolen art to advertise a tool that directly benefits him. Not a lawyer but that looks like a slam-dunk of a case,' one alleged.
Others, however, did not believe Mr Brooks' case was strong enough.
Mr Brooks told Daily Mail Australia it was unlikely he would pursue that avenue.
Musk's Grok engine has been the subject of controversy in recent weeks as allegations of censorship plagued the platform.
Grok is a conversational AI developed by xAI, founded by Musk. The engine can access contemporary information over the web and on its native platform, X.
Its use has been predicated on its ability to answer questions typically avoided by other platforms, and can be accessed at varying degrees of utility through free and paid accounts.
However, in recent weeks the chatbot suffered a 'programming error' which made it sceptical of Holocaust figures and repeatedly made claims about a 'white genocide' in South Africa.
In May, researchers revealed AI models had begun disobeying human instruction.
ChatGPT's latest AI model refused to switch itself off, they claimed.
Experts said they gave the AI system a clear command but the o3 model, developed by OpenAI and described as the 'smartest and most capable to date', tampered with its computer code to avoid an automatic shutdown.
