What's new

AI Image Generation Software continues to improve

Click For Summary
SUMMARY

OpenAI's latest image generation feature integrated into ChatGPT 4.0 offers an intuitive interface and impressive understanding of complex prompts, although it may not achieve the highest fidelity in image quality. Users report that the tool effectively generates detailed images based on specific instructions, allowing for significant customization. This advancement raises important questions about the nature of evidence and authenticity in digital content creation.

PREREQUISITES
  • Understanding of AI image generation concepts
  • Familiarity with ChatGPT 4.0 features
  • Basic knowledge of prompt engineering for image generation
  • Awareness of ethical implications in AI-generated content
NEXT STEPS
  • Explore advanced prompt techniques for ChatGPT image generation
  • Research the ethical considerations of AI-generated imagery
  • Learn about the capabilities and limitations of Midjourney
  • Investigate the impact of AI on digital evidence and authenticity
USEFUL FOR

This discussion is beneficial for digital artists, content creators, AI enthusiasts, and anyone interested in the implications of AI in media and communication.

Messages
297
Reaction score
215
OpenAI's new image gen built into ChatGPT doesn't have the very best fidelity, but it has an incredibly easy to use interface, and the image generation very easily understands what you are trying to get it to produce, even when it is relatively complex. It also does not have many restrictions on targets.

Since this is the worst this technology will ever be for the rest of your lives, it will almost certainly call into question what constitutes evidence in the near future.

Screenshot_20250326_161151_Gallery.webp
Screenshot_20250326_161113_Gallery.webp

Screenshot_20250326_161132_Gallery.webp
Screenshot_20250326_161052_Gallery.webp


All images generated by ChatGPT image generation.
 
Last edited:
The image generator on ChatGPT 40 is very good. What's really new about it is its ability to tweak the image based on your input. Here's an example of what I got it to generate this week along with the prompts that I used.

Generate an image of a bunch of traitors entering the white house while leaving a trail of destruction behind them. Make it look apocalyptic.
1743758950318.png

Not quite so dark and have religious people on the sides fawning over them and prostrating themselves.
1743758993521.png


Now add some people in dunce caps along with other representations of completely moronic people. Put a dunce cap on the priest to the left.
1743759055241.png


Except for the central traitor, put a fat, yellow streak down the backs of the other traitors.
1743759102223.png
 
I played with Midjourney for a bit early on, but I'm happy to see it native to GPT4o now. It's a bit slow, but it's still pretty good and convenient.
 

Liberal Democracy Values

  • Free and Fair Elections
  • Rule of Law
  • Separation of Powers & Checks and Balances
  • Protection of Civil Liberties and Human Rights
  • Pluralism & Political Competition
  • Independent Media & Free Press
  • Open Civil Society

Community Values

  • Civility
  • Productivity
  • Good Faith Debate
  • Evidence Based Debate
  • Transparency
  • Integrity

Community Motto

"It is the responsibility of intellectuals to speak the truth and expose lies." - Noam Chomsky
Back
Top