Bug Reports

When creating the text A bold and Angular Gold Font with styles from medeval Celtic armies Simple, the AI generated object out side of the text area, I can never imagine someone wanting this even if it was leaves it should still be in th boundry area.

Alot of images are distorted. The faces are not clear, the fingers are weird. For example, if you try the prompt: Goddess Parvati in a pond discussing with her sisters- It's impossible to get good images. I hope Adobe fixes this issues of distorted image generation.

 Need to be able to import photoshop files more easily. The export as a svg file in photoshop is not working for many. Does it have to be a specific format to do so. I need to work with .psb files since my files are so large - photos printed 70' long x 48' high - yes thats feet. https://community.adobe.com/t5/photoshop-ecosystem-discussions/p-how-to-export-to-svg-from-photoshop/td-p/12208976#:~:text=Hi%20there%2C%20If%20you%20are,in%20Export%20As%20Dialog%20Box.

letters in text effects can be blown out like bad erasure

Some words are disabled for example weed as a violation of the safetey hower you can still write canabis and get the same results 

Hello. I posted some work on Firefly and it came to my attention that the prompt info is disclosed. As a poet, I tend to keep my prompt info close to my chest. I feel like it's a trade secret, as others can create a bot and continue using it to generate work. Is there a way to hide prompt info? 

Hola, hace dias pude generar en la version beta, todo bien, ahora intento y me sale este error siempre   You're offline Please check your connection and try again.   no he podido generar ninguna mas desde hace dias por el mismo error, ya fui aceptado para el beta, pero ya no he podido generar mas.

"Add human existence" What is the problem with this sentence?Thanks

The application crashes whenever I press the button to download a selected image.

One of the pictures failed to load and after reloading it still failed.   

Favourited images not appearing on all devices.

1. Beta2. Lap top; MacOS Monterey V12.6.5; chrome browser3. Several variations of the following prompt "happy middle aged woman sitting, shot from bottom of foot and upward"; Content type: photo; Composition: shot from below; Aspect ratio: 1:1, 16:94. Everything except for the face is great. The face is distorted--warped, split in half, features in the wrong position, etc.

Favorite images saved on the iphone is not shown in the website with the same account.So if I favorite somehting on my iphone it will show up only on my iphone and not on my laptop.

I saved a bunch of favorites on my computer, then logged into my other personal laptop and none of my favorites are saved.

I found an interesting glitch. Apparently Firefly has a problem with orange peels. "Can't loadWe encountered a safety issue with this image. Please try again."  

Freeze the browser after couple tries.

Hello,I had come here from Firefly, wanting to report an issue with the generated output. But, I guess that can wait given it's still in Beta and I am sure the team is working hard on that front.While writing this, I have decided that I will rather speak on this very page on which I am reporting a bug. Why can't you capture my feedback within the app, take the screenshot there itself, and let me mark the area that I am having a problem with?... You redirect me to a blank page expecting the user to describe the problem when you made me lose the context by redirecting me to a completely new browser tab. On top of that, you want me to take a screenshot of the app, save it on my computer, and upload it here? Team, please, this method is archaic. Please look at how Youtube and other Google products capture feedback. 

I tried several prompts and each time only got 4 error messages like this:"Can't loadThis image produced an error. Please try again." I'll try a different device and browser to see if that helps.

I wanted to share my feedback regarding your AI tool that generates images from text. While I appreciate the potential of this technology, I have noticed that the tool still has a lot of errors and limitations in replicating human-level creativity and imagination.In my experience, the generated images often lack the contextual understanding and cultural knowledge that humans possess. Additionally, the tool may produce inaccurate or inappropriate results due to biases and limitations in the underlying datasets used to train the AI model.I believe that there is a lot of room for improvement in this area, and I encourage your team to continue to work towards developing more advanced and accurate AI tools for generating images. As a customer, I would appreciate seeing advancements in this technology that can match or exceed the quality of human-generated content.Thank you for considering my feedback, and I look forward to seeing the progress of your AI tools in the future. For Example

Hi All, Just noticed that my Favourites are not syncing from my iPhone to my Mac Pro when using Safari and both are logged in. Is this a known issue? or is it just not possible at this time? Kind Regards Jamie Dockerty Hardware Specs:iPhone OS 16.4.1(a)Mac pro Trashcan Late 2013Monterey 12.6.4Safari Version 16.4.1 

Font ain't changing in Text effect. It's always generating serif fonts.

Hi! I generated, downloaded, read the Layer saying "we write Credentials into metadata", but the file didn´t contain any. I copy the XMP content here. I haven´t overseen any info, right?Best, Alex{"ExifTool": {"ExifToolVersion": 12.0},"System": {"FileName": "Firefly human brain, light from bottom, created with black mamba snakes, heads visible 61648.jpg","FileSize": 815568.0,"FileModifyDate": "2023-05-15T18:03:29.000+00:00","FileAccessDate": "2023-05-15T18:05:53.000+00:00","FileInodeChangeDate": "2023-05-15T18:05:53.000+00:00","FilePermissions": 100644.0},"File": {"FileType": "JPEG","FileTypeExtension": "JPG","MIMEType": "image/jpeg","ImageWidth": 1024.0,"ImageHeight": 1024.0,"EncodingProcess": 0.0,"BitsPerSample": 8.0,"ColorComponents": 3.0,"YCbCrSubSampling": "1 1"},"JFIF": {"JFIFVersion": "1 1","ResolutionUnit": 0.0,"XResolution": 1.0,"YResolution": 1.0},"JUMBF": {"JUMDType": "6332706100110010800000aa00389b71","JUMDLabel": "c2pa"},"ICC-header": {"ProfileCMMType": "","ProfileVer

Sometimes I can't download generated image, the button simply dosen't work. Along with that, usually I can't open the image anymore. It appear instead "Selected: Variation {k} of 4". Howevere, if I use the "show similar" button to generate 3 new images similar to the original, I am now able to open and download the original image.

R.2iyjukr.lhoix0l6

Could not generate image with proper text for qoutes. Prompt: Build the dream and the dream will build you