Developers Can Now Tap OpenAI’s Image Model Through API 🎨

Welcome to another edition of Horizon AI,

OpenAI announced today that it is finally bringing its highly popular image generation feature in ChatGPT to the API.

Let’s jump into it!

Read Time: 4.5 min

Here's what's new today in the Horizon AI

• OpenAI Makes New Image Generator Available for Developers via API

• Adobe Expands AI Offerings with New Firefly Image Models and Creative Cloud Updates

• AI Tutorial: How to apply Content Credentials metadata to protect your images from AI training

• AI Tools to check out

• The Latest in AI and Tech 💡

• AI Findings/Resources

AI News

OPENAI

OpenAI Makes New Image Generator Available for Developers via API

The company announced that gpt-Image-1, the natively multimodal model powering ChatGPT's image creation tool, is now available globally via API, allowing developers to integrate it into their apps and services.

Details:

• OpenAI’s upgraded image generator has proven extremely popular, with over 130 million ChatGPT users creating more than 700 million images in just the first week of the tool's availability.

• The model can create images in different styles, follow custom guidelines, leverage world knowledge, and render text.

• Developers can also control moderation sensitivity, which can be set to “auto” for standard filtering or “low” for less restrictive filtering.

• Pricing is $5 per million input tokens for text, $10 per million input tokens for images, and $40 per million output tokens for images. (Approximately 2 cents, 7 cents, and 19 cents per generated image for low-, medium-, and high-quality square images, respectively.)

• OpenAI says that companies like Adobe, Airtable, Wix, Instacart, GoDaddy, Canva, and Figma are already using or testing gpt-Image-1.

This API release strengthens OpenAI's position in the developer market and could open up new revenue streams from its most popular features.

Source

ADOBE

Adobe Expands AI Offerings with New Firefly Image Models and Creative Cloud Updates

Adobe is stepping up its AI game with the release of two new Firefly image models, along with updates to its Creative Cloud apps like Photoshop and Illustrator.

New Firefly Models:

• Firefly Image Model 4: Adobe’s fastest and most realistic image generator yet, supporting up to 2K resolution with more control over styles, formats, and camera angles.

• Firefly Image Model 4 Ultra: Designed for complex and highly detailed tasks, perfect for creating intricate scenes.

Expanded Features:

• Firefly’s text-to-video and text-to-vector models are now available in the Firefly web app alongside the new image models.

• The new Firefly Boards feature, a collaborative AI moodboarding tool, has launched in beta.

• A Firefly mobile app is “coming soon” for iOS and Android.

Third-Party AI Integration:

• Users can now experiment with OpenAI’s new GPT image model, Google’s Imagen 3, Veo 2, and Flux 1.1 Pro directly in the Firefly web app, with support for Luma, Pika, Runway, Fal.ai, and Ideogram “coming soon.”

Creative Cloud Updates:

• Illustrator’s Generative Shape Fill and Text-to-Pattern tools are now generally available, following their beta launch last year.

• Photoshop introduces smarter color adjustments, enhanced selection tools for fine details like hair and facial features, and an improved Actions panel for tailored editing suggestions.

Source

AI Tutorial

How to apply Content Credentials metadata to protect your images from AI training

Adobe has just launched its new free Adobe Content Authenticity app, allowing creators to apply invisible, tamper-resistant Content Credentials to images and photographs to help identify who owns them.

1. Go to the Content Authenticity web app and log in with your Adobe account.

2. You can batch-apply Content Credentials to up to 50 files at a time.

For now, only JPEG and PNG files are supported, but Adobe says that support for larger files and additional media, like video and audio, is "coming soon."

3. Select the information to attach to your work, such as your name and social media.

Here you can choose whether you want your work used in AI training and usage.

Users can only express their preferences—while some companies may ignore them, the hope is they’ll still respect the creator’s intent. Since these credentials stay with the work permanently (even if someone takes a screenshot), user preferences could become more impactful as laws and opt-out regulations evolve.

4. To double-check an image’s authenticity and learn more about its Content Credentials, use the Inspect tool. Just drag and drop the content you want to learn more about.

Note: The tool is still in beta and currently free for everyone. You may be placed on a waitlist initially—just wait until access is granted.

AI Tools to check out

⚡ BuzzClip: Generate viral TikTok AI UGCs in 60 seconds.

🚀 Velocity: A Prompt co-pilot that turns unclear instructions into optimized, AI-ready prompts in one click.

✅ AutoDiagram: Transform ideas into professional diagrams with AI.

💤 Remy AI: Sleep & health coach.

✨ Remento: Turn memories into a keepsake book. No writing required.

AI Findings/Resources

📐 How might AI affect architects? A Yale expert weighs in

🔮 What's next for AI at DeepMind

👉 Open source technology in the age of AI - McKinsey & Company

🎬 Films made with AI can win Oscars, Academy says

The latest in AI and Tech

OpenAI eyes $174B in sales by 2030

Total revenue is projected to surge to $125 billion in 2029 and $174 billion the following year, from $3.7 billion in 2024.

Ray-Ban Meta Glasses get new features and increased availability

Meta has announced several new features, a couple of new styles, and expanded access to seven additional European countries.

Anthropic anticipates fully AI employees arriving as soon as next year

AI-powered virtual employees could begin operating within corporate networks as early as next year, according to Anthropic’s Chief Information Security Officer, Jason Clinton.

Trump signs AI education order to train K-12 students

President Donald Trump signed an executive order directing the integration of AI education into U.S. K–12 schools.

The order mandates the Departments of Education and Labor to develop AI-focused curricula and certification programs to equip students with skills for navigating emerging technologies and encourages the incorporation of AI tools in teacher training.

That’s a wrap!

Thanks for sticking with us to the end! Let’s stay connected on LinkedIn and Twitter.

Not subscribed yet? Sign up here and send it to a colleague or friend!

See you in our next edition!

Gina 👩🏻‍💻