👨‍🚀 6 new ChatGPT hacks for viral images

Hello AInauts,

Welcome to the latest issue of your favorite newsletter!

Today with a very practical focus on image generation and editing with the latest tools. We have tested them all - with a few surprises…

• 👀 New image editors ChatGPT vs. Grok vs. Gemini etc. in our tests

• 👌 6 clever ChatGPT hacks for your pictures

• 👣 AI-Fun: Is this what Bigfoot looks like?

Here we go!

👀 Practical test: ChatGPT vs. Grok vs. Gemini etc. - Which image editor really delivers what it promises?

We have already talked in detail about the cool possibilities of the new Google models over the last few days (Newsflash: the new Gemini 2.5 Pro is now available for all users to test here).

But ChatGPT and Grok have also introduced new updates that can be used to generate and edit images. The ChatGPT update in particular has caused viral hits and a frenzy.

Naturally, we wanted to find out who was really ahead of the game - and carried out lots of tests over the weekend. For good measure, we also included the new Ideogram V3 model, Flux and the viral image generator Reve. Midjourney can only be used with a paid account and is not included.

Test #1: Simpsons-style image generation with text & persona

Our test criterion: Who can come up the most absurd Simpsons-style Trump meme? Spoiler: All have passed 😁 ...

ChatGPT and Google Gemini are convincing, although Gemini is much faster than ChatGPT.

Only ChatGPT did the hint from the prompt to "hold the instruction manual upside down" to some extent correctly. Gemini surprised us with a Homer robot, and Reve and Ideogram also made a good impression.

Grok and Flux (via Freepik), on the other hand, did not convince us - we had to make half a dozen attempts to get the text and persona to fit halfway, and find the result rather poor...

Here is the prompt used:

Image: Depict a Simpsons-style scene where Donald Trump is attempting to assemble a robot kit labeled 'AInauten.com AI-Bot.' The partially assembled robot looks comically dysfunctional. Trump is holding the instruction manual upside down, with a frustrated expression. A speech bubble says: 'Made in China?' The background features a typical living room setting with tools and robot parts scattered around. The image should be in 16:9 format.

Test #2: Editing image style and text

Let's move on to the second test: editing images with voice commands. We will limit ourselves to ChatGPT, Gemini, Grok and Reve.

This example makes it immediately clear who has the edge here! ChatGPT impresses with error-free text, identical image composition and realistic photo style. Newcomer Reve also gets off to a strong start.

Gemini struggles to make the whole scene realistic, and Grok doesn't seem coherent.

We had to make a few attempts here. Gemini changed the composition, Grok became dyslexic and ChatGPT refused to work after half of the image... But in the end, it worked (if necessary, a context reset helps: simply copy the image into a new chat and make the adjustment there).

Make the image as a hyper-realistic photograph
Change the text on the manual to "Anleitung für Donald" and "Top Secret"

👌 6 clever ChatGPT hacks for your pictures

ChatGPT has won us over. But that was just the beginning, because there are a few more helpful features since the latest update. (And an oopsie... custom GPTs that supported image generation with DALL-E are going nowhere).

Pro Tip #1: ChatGPT Edit function is your ace in the hole

Simply click on the image, then select the brush, highlight the area you want to change, enter a prompt and ... boom ... the image is adjusted! A real killer feature that very few people know about and use it.

Pro Tip #2: Share your chat including picture(s) with others

Yes, you read that right. ChatGPT finally allows you to share chats that contain images - previously only text-only chats were available for sharing. This makes it easier for others to further develop your prompts and creations. What doesn't work is sharing chats where you have uploaded images yourself.

Pro tip #3: Create social media texts and images with a prompt

You can also create the perfect social media post with a one-shot prompt. This works in no time at all, for example with this template:

1. Create a suitable, crisp social media text on the topic below.
2. Generate an image for me.
IMPORTANT: everything in English, you-form - make it go viral!

The image will look like this, the text generated for the post can be found here.

Pro Tip #4: Create GIFs directly in ChatGPT

Finally, we have a tip from Prompt Magician Pliny - you can create GIFs directly in ChatGPT, instructions here.

Pro Tip #5: Create photos and matching videos directly in Sora

In OpenAI's video generator Sora, there is also a new section for creating images, with lots of examples - and the option to use filters and create two image versions at once.

If you run into a time limit with ChatGPT and have to wait, you can simply switch to Sora and create images directly. What is missing there, however, is the option to customize them (see Pro Tip #1).

But instead, you have the option of creating a video directly from an image with just one click - simply click on "Create Video" in the detail view below.

Pro Tip #6: Create transparent PNGs in ChatGPT, in your brand colors

If you want a transparent PNG, just include it in your prompt. And if you want it in certain colors, then add the hex codes.

Creating and customizing images is as simple as chatting using GPT‑4o—just describe what you need, including any specifics like aspect ratio, exact colors using hex codes, or a transparent background. pic.twitter.com/fnen1aDjNZ

— OpenAI (@OpenAI) March 25, 2025

Our take: ChatGPT kills startups once again - and offers Photoshop-level possibilities for everyone

With every significant update of ChatGPT, new functions are added that enable us average consumers to do things that we previously had to hire a specialist for.

What can I do with ChatGPT?

Whether it's designing flyers, websites, products, mockups or editing existing photos: ChatGPT can support you here. Only your creativity sets the limits here!

It can customize polaroids, paintings, ads, more ads, ads, more ads, thumbnails, comic book covers, comics, mangas, Memes, more memes, Pixar memes, customize image styles, Ghibli memes, Ghibli spirals, consistent characters, games, Game sprite sheets, movie posters, landing pages, user interfaces, mobile apps UIs, iMessage screenshots, IDs, receipts, security camera images, infographics, more infographics, mindmaps, maps, recipes, logos, social media profiles etc. can be created.

The highlights on X and in the Sora stream also offer plenty of inspiration!

Can I now do everything with ChatGPT and co. - and quit Canva etc.?

In general, the restrictions on the leading providers have been significantly relaxed, giving users more flexibility in image generation - which will undoubtedly lead to a wave of lawsuits against OpenAI (analogous to MidJourney and Stable Diffusion - the proceedings will begin in September 2026).

OpenAI's Joanne Jang explains the internal considerations and quotes aptly: "Ships are safest in port; the safest model is the one that rejects everything. But that's not what ships or models are for."

The legitimate question arises as to what else you need a subscription to Canva or Photoshop etc. for. Sure, there may be specialized features for CI/CD compliance, many templates to choose from, or it is easier to share and edit content in a team.

But for the occasional user, ChatGPT will definitely be an extreme relief in the daily process. And other image generation tools such as Midjourney will also have to do some soul-searching ...

Why is ChatGPT so much better here than its competitors?

The reason lies in the image model used. While the entire world uses a diffusion model, the GPT4o model uses an autoregressive approach. This enables much more precise images and coherent texts - but also requires more time.

And if you don't want to wait until ChatGPT offers the feature for free users, you can also achieve good results in image creation and editing with Gemini or Reve. Grok, on the other hand, disappointed us.

Last but not least, a shout-out to Google: If Gemini hadn't released its own first-class image generator, we would probably have to wait a long time for this great update!

👣 AI-Fun: Is this what Bigfoot looks like?

Phew, it was a long edition today - so let's finish with something a little more light-hearted. We found an example of an upscaled Bigfoot photo on Reddit - and naturally wanted to try it out with the aforementioned image editors…

We made it! But no need to be sad. The AInauts will be back soon, with new stuff for you.

Reto & Fabian from the AInauts

P.S.: Follow us on social media - that motivates us to keep going 😁!
X, LinkedIn, Facebook, Insta, YouTube, TikTok

Your feedback is essential for us. We read EVERY comment and feedback, just respond to this email. Tell us what was (not) good and what is interesting for YOU.