0 comments
get your own sh*t!
Try to distinguish between a genuine Studio Ghibli image and something thrown together in 30 seconds on ChatGPT, we dare you.
Generative AI platforms have become so fast, polished, and culturally omnivorous that any visual style can now be coopted. It’s good clean fun when making crude memes for the group chat, but the implications for people who earn a crust in creative industries are grim.
Why buy that canvas featuring one of your favourite works for £500, when AI can closely replicate it and link you to Vistaprint before the kettle has even boiled? Oh, you’ve put watermarks on your image. I’ll have it without copping the licence fee, thanks.
That sort of cultural mugging is exactly why the UK government is trying to tighten the leash. As part of a wider copyright reform package, ministers have been meeting in recent weeks to discuss the possibility of labelling AI generated content online. A digital taskforce is now working in the background to establish how this can be achieved: badges on posts, small lines of text in captions, ‘more info’ notices, a hidden technical marker in metadata. Those sorts of things.
We know that the UK has been particularly tetchy over AI being used for sexualised deepfakes or spreading misinformation, but this is the first major sign that the government is taking the plight of artists seriously.
As the 10,000 strong Don’t Steal This Book protests reminded us at the London Book Fair, creators have unionised and lobbied against their works being stolen for years now – and they’re seriously up against it.
As machine learning systems continue to hoover up any and all online content, the need for permission has eroded completely. Artists and authors have pushed for a system that requires AI companies to have explicit consent for training data, not an opt-out model that puts the onus back on the artists to fend off digital scavengers. Thankfully, the government is planning on ruling out that possibility, or so the grapevine says.
View this post on Instagram
Make no mistake, technology minister Liz Kendall is all in on AI, recently reiterating that the sector is growing 23 times faster than the rest of Britain’s economy. Given the technology is clearly taking precedence, we’d be foolish to expect a legislative catch-all to liberate artists immediately.
What is positive, however, is that the creator economy has finally been acknowledged and the burden of responsibility heaped back onto the nation’s leadership.
Labelling all AI generated material online may sound like the government biting off more than it can chew, but hopefully artists can come out of the situation more favourably than expected.
❤️ love audit
Gen Z Is Particularly Weird About Relationship Age Gaps. Here’s Why – huffpost
HuffPost digs into why Gen Z tends to get a bit twitchy about age-gap relationships, especially when one person is barely into adulthood and the other ditched their first wife 15 years prior. The article argues that its less about the number itself and more about what the gap represents, mainly differences in maturity, experience, and what it potentially reflects in the – insert buzzword – power dynamics. Younger people are more likely to question these aspects of relationships anyway, so the wider the gap becomes, the higher the suspicion grows that something may be amiss. Given we grew up online and are overly familiar with how manipulation and coercion can play out in extreme and dramatic ways, it’s no surprise the benefit of the doubt isn’t given straight away. This is especially true if your model girlfriend wasn’t even around to see the original Spiderman trilogy, Tobey.
https://www.youtube.com/shorts/c_Z6U_lN4j8?feature=share
Gen Z is dating less. The result is one of the most unprepared workforces – fortune
Our generation being useless at dating is apparently affecting our work lives also, according to psychologist Tessa West who had some interesting takes in a recent Fortune article. She said that the need to navigate awkward, compromising, and emotionally charged moments while dating inadvertently flexes muscles needed to succeed in the professional world. The piece reports that only about 56% of Gen Z enter adulthood having had a romantic relationship, compared with 75% of older generations, and uses that as a jumping-off point for a broader argument about social skills. Is it a bit of a stretch, or is our reluctance (or perhaps inability) to embrace real world intimacy dulling the instincts and social development needed to get anywhere at work? Could the work/life scale more interconnected than we realise, or is Tessa waffling?
🤖 ai asthenia
Crimson Desert developers apologise for using AI art – thred
‘Ahh sh*t, here we go again.’ The use of AI generated assets in a videogame has, once again, ignited the ire of players. Pearl Abyss can’t really be accused of laziness for its flagship title Crimson Desert, given how technologically masterful it is – if lacking a little in narrative richness, yada yada. However, the inclusion of AI generated 2D assets such as in game paintings has annoyed and disappointed the community. The developers were quick to put out a statement saying that leaving the AI bits in was due to oversight, with said items intended only to be used as placeholders during development. Some are giving the benefit of the doubt, and the more cynical bunch are treating it as a creative and ethical choice. It’s trendy to hate on Call of Duty, but if others follow the same path, they’ll hear about it too. Thankfully, the devs are rolling out a patch which will soon nuke the low-quality AI objects from the stunningly realised world of Pywel.
BMG sues Anthropic for using Bruno Mars, Rolling Stones lyrics in AI training – reuters
Reuters reports that BMG has sued Anthropic in California federal court, alleging the company used copyrighted lyrics from artists including Bruno Mars, Ariana Grande, and the Rolling Stones to train Claude without permission. As we said at the start real implications of the newsletter: real implications! The complaint says the AI copied lyrics from unauthorised sites and infringed 493 separate complaints. While the legalities of AI and machine learning remain up in the air, Anthropic has argued that its training methods are fair game, given copyrighted material isn’t technically off-the-table. It just goes to show, aversion to generated works isn’t just coming from disgruntled graphic designers or writers. The cast of your latest Spotify playlist also wants a word about having their rights reserved.
