When I first got access to DALL-E, it was not only an amazing life experience but since I got in early (thank you again OpenAI team), I was given access to the tool in a pretty much unlimited capacity. This meant, specifically, I would receive 9 image results at a time, unlimited prompt lookups, unlimited variations, and unlimited edits/inpainting. I am still grateful for the opportunity.
What I immediately found was that after several prompt lookups in the platform, I not only developed an intuition for the modelβs capabilities, but I felt my imagination and a space of possibilities growing faster and wider than ever before around me. I was constantly thinking of prompt ideas and had several lists going at the same time, exploring all of them as fast as I could type!
The whole experience was visceral and I could feel my creative abilities running a mile a minute. I didnβt sleep for 2 weeks straight and I found myself using the DALL-E app even on my phone after just finishing using it on my desktop.
However, since the DALL-E 2 research beta has seen several quotas and restrictions introduced, I have found myself using DALL-E a lot less. Specifically, the 50/prompts per day quota must have seen me using the product maybe a tenth as much as I used to.
I spent some time last week thinking about why this was. On one hand, I was grateful more people would get access to try it too. On the other hand, I thought back to some of my favourite DALL-E 2 renders:
β¦ and I realized they were all the result of so many later stage prompts which I entered into DALL-E. They were outputted very well past the 50 prompt per day limit. At some point, you are so deep into prompt lookups and iterating so fast, you canβt necessarily even trace back how you got to where you are. Youβre rapidly exploring new territory, iterating on prompts/ideas quickly, and at the upper limits of your imagination and mental visualization abilities. All of these above images were not what I had initially set out to make when I started entering prompts into DALL-E, but through a longer process of iteration and discovery, I was able to discover them. What I believe is my best and most original work which I also enjoyed making the most.
However, since the quota, I have not been able to reach this important state of mind. At first, I struggled to find an adequate word for this stage, but then I realized it already exists for other creative disciplines - artists, musicians, programmers, writers, all describe the importance of reaching some kind of βflow stateβ where they can work at their peak creative levels.
Why is why, a week or so ago, I tweeted this:
Not being able to reach a flow state is a serious problem for the future success of DALL-E 2. Not reaching a flow state will reduce the quality of your work, the originality of your ideas, and limits you from being able to practice your craft. I actually think this quota is dangerous because most creatives create in the first place, so that they can reach this elusive state of mind. This βhighβ is what creatives are chasing in the first place and attaches them deeply to the craft itself.
Without being able to get into a flow state, you would never be able to get into the Saxophone, Adobe Photoshop, or your favourite programming language. The flow state allows you to reach the height of your abilities, escape reality, it provides therapeutic value, releases endorphins, and offers some kind of other worldly mental experience which all creatives yearn for. I think the flow state is key to developing a real attachment to any craft.
In a previous podcast episode, at two different parts of the discussion, I argued for a very thoughtful approach to DALL-Eβs pricing and quota model. However, one of the main things I suggested was giving new DALL-E users far more generous access and quotas, so they too, can develop an intuition for the product and reach some kind of flow state. My belief was that, after experiencing an initial flow state, this would create a DALL-E user for life.
For the sake of the future of the DALL-E 2 product, I think itβs critical that OpenAI consider the effects of their quotas and pricing on the usersβ abilities to meaningfully reach their own flow state. It is to critical to understand the data and analytics around this flow state inflection point, when do users reach this flow state? How can they get there faster? Which pricing and quota structures takes away from usersβ opportunities to reach their flow state? By focusing on the effects on the end usersβ flow state, I think we could improve the satisfaction, quality output, originality, retention, virality, and product value of DALL-E 2 and create a pricing and quota structure which works for everyone.
I've only been using Dalle-2 for afew hours, but that is exactly the feeling I got. Hopefully they will look to Mid-journey and evolve a similar pricing structure