GPT-4o Is in Microsoft Copilot Now: Here’s What You Can Do With It

Why Trust Techopedia

Just weeks after OpenAI’s big GPT-4o update gave ChatGPT new image generation powers, Microsoft is bringing that same functionality to its own Copilot AI assistant.

Now baked into Microsoft 365 apps like Word, Outlook, and Excel, the feature lets you create images from text prompts right inside the tools many of us already use daily. And according to Microsoft, apart from making images from scratch, you can also use Copilot to tweak and refine existing ones, with the help of GPT-4o’s upgraded visual capabilities.

This move isn’t surprising considering that Microsoft has spent years and billions investing in OpenAI, weaving its tech into the heart of its software. What’s new is how accessible that tech is becoming to end users.

In this guide, we’ll walk you through everything you need to know about the new Copilot image generation feature with the GPT-4o model and how to use it.

Key Takeaways

  • Microsoft has added GPT-4o’s image generation capabilities to Copilot, making it available in Word, Excel, Outlook, and other Microsoft 365 apps.
  • Users can now create images from text prompts, edit existing images, and change styles directly within Copilot.
  • The GPT-4o model allows for more detailed, photorealistic images with realistic lighting, depth, and human expression.
  • You can also upload your own images and use them as a starting point to generate new visuals.
  • This feature was previously limited to enterprise users but is now available to the general public.
  • Microsoft’s integration of GPT-4o signals a deeper push to compete with AI tools like ChatGPT and Google Gemini.

What Is GPT-4o & Why Does It Matter?

GPT-4o is OpenAI’s latest multimodal model capable of understanding and generating text, images, and audio in real time. Unlike its predecessors, GPT-4o has been trained to process different types of input, which enables it to deliver natural, fluid responses across formats.

According to Microsoft, upgrading Copilot’s image generation with ChatGPT-4o means that users can now create images with richer details and compositions. In a press statement in May, Microsoft said:

“Copilot also does a much better job of refining images based on a previous image, and now you can even upload your own image to use as a starting point, giving you more creative control and flexibility.”

It is worth noting that Microsoft first rolled out these GPT-4o image generation tools to enterprise users via Microsoft 365 Copilot last month. However, these same features are now available to the general public through the consumer version of Microsoft Copilot.

For professionals and casual users, this could mean a lot, since the update will likely reduce the time it takes to create visual content.

When to Use GPT-4o Image Generation in Copilot

Microsoft revealed in a YouTube post that GPT-4o image generation in Copilot will bring creativity to life. Based on the video, you may try using it if you want to accomplish the following:

Render accurate, readable text
You can use the GPT-4o image generator in Copilot if you need to embed clear, purposeful text directly into an image without post-editing. This is helpful if you’re working on content where topography is central to the message or design.
Edit what you’ve created
Let’s say you already have an AI-generated image in Copilot but want to make changes without starting over. The new GPT-4o image generator allows you to achieve this without losing your initial efforts.
Create complex, layered images
Another way to use the GPT-4o image generator in Copilot is by prompting it with more detailed ideas. This can be useful when you’re trying to capture something that has multiple elements or a lot of visual depth. You can use it for concept art, educational diagrams, fantasy scenes, or anything that needs different subjects, styles, or perspectives, all brought together in one image.
Change the style of an existing image
Copilot also comes in handy when you have an image you like but want to see it in a completely different style. Maybe you’re aiming to reach a new audience or adapt the same idea for different platforms; you can keep the overall layout or composition and simply shift the aesthetic, which is useful for staying consistent while still experimenting creatively.
Make realistic images
If you need to create AI images that closely mimic real photography, then the GPT-4o image generator in Copilot can also be helpful. Microsoft suggests the new GPT-4o update in Copilot can help you generate realistic images with accurate lighting, depth, shadows, materials, and human expressions for use in different settings.

Step-by-Step Guide to Using GPT-4o Image Generation in Copilot

If you’re new to Copilot, follow these simple steps to get started with using the new image generator in the AI assistant.

  1. Go to copilot.microsoft.com or open Copilot within Microsoft 365 apps (like Word or PowerPoint). If you’re using a mobile, you can download the Copilot standalone app from the Google Play or Apple App Store.
  2. Once you’re in the Microsoft 365 Copilot app, navigate to the “create an image” button.
  3. Decide what to do by choosing your image task.
  4. On the Copilot chat, write a clear prompt that is specific and descriptive.
  5. Generate an image by hitting the Enter key, and then wait for the image to appear.
  6. Edit or refine the image with more prompts if you want to make some changes, and Copilot will apply these changes directly to the existing image.
  7. Click the download icon to save the image locally. Or, if you’re in a Microsoft 365 app, insert it directly into your document or slide.

The Bottom Line

Microsoft’s ongoing investment in OpenAI is starting to pay off. With GPT-4o, Copilot can now generate images more quickly and with greater detail, offering a smoother creative experience.

The update is no longer limited to enterprise users, which means more people can try it out. This move positions Microsoft more firmly in the race alongside tools like ChatGPT and Google Gemini.

However, it’s difficult to say what kind of impact that will have just yet. But the next few months will likely reveal how users respond.

FAQs

What image generator does Copilot use?

Can Copilot image generator create Ghibli memes?

Related Reading

Related Terms

Advertisements
Franklin Okeke
Technology Journalist
Franklin Okeke
Technology Journalist

Franklin Okeke is an author and tech journalist with over seven years of IT experience. Coming from a software development background, his writing spans cybersecurity, AI, cloud computing, IoT, and software development. In addition to pursuing a Master's degree in Cybersecurity & Human Factors from Bournemouth University, Franklin has two published books and four academic papers to his name. Apart from Techopedia, his writing has been featured in tech publications such as TechRepublic, The Register, Computing, TechInformed, Moonlock, and other top technology publications. When he is not reading or writing, Franklin trains at a boxing gym and plays the piano.

Advertisements