If GPT Images v2 made these, I think we may have finally bridged the gap between frontier AI slop and actual art, or at least we are starting to see a real semblance of tasteful art.
Midjourney has been the best at this for a while and has been able to produce actual art for some time, but I think it is safe to say GPT is now entering that category too. I believed DALL·E 2 was art back in the day as well, but there was still a kind of slop in most frontier models, even as the images technically got higher quality. It was easier to tell that they were mostly just amalgamating their sources rather than expressing a vision of their own.
This is the first time I have seen one of these models actually creatively induce the feeling of an art style in what it wants to create, rather than just approximating references at a higher level of fidelity
🚨 Holy shxt… The rules of AI image generation just completely changed.
A GPT-based model currently testing in the arena under the bizarre name 'duct-tape' is turning the global AI community upside down. What exactly did they feed this thing?
→ Native-level text generation with zero awkwardness. → Chilling consistency maintained down to the pixel. → Overwhelming illustration quality ready for immediate commercial use.
"Is it just downloading photos from the internet?" That was every tester's first reaction. It's that unbelievable.
A game-changer that instantly crumbles Nano Banana Pro's dominance has arrived. A lot of people might need to start packing their desks again.
The 10 images below were 100% drawn by AI. See for yourself 🧵