ChatGPT Images 2.0: a test of branding, text, and infographics — impressive, but not without errors

ChatGPT Images 2.0 is the largest upgrade to image generation in ChatGPT in recent times. A ZDNet journalist conducted a series of practical tests and found: the model makes a major leap forward, especially in working with text, branding, and infographics — though not without errors.

What Changed in Images 2.0

The main pain point of AI image generation for several years has been text. Previous generations of models either completely avoided text in frames or produced illegible gibberish from pseudo-characters. ChatGPT Images 2.0 handles this fundamentally differently: headings, graph captions, short blocks of text in infographics — everything comes out readable and positioned in the right places.

The second major shift is working with branding. The model understands requests to create visuals with specific color codes, designated typography, and brand elements. This opens a new use case — rapid prototyping of marketing materials without designer involvement.

How the Test Went

The journalist systematically tested Images 2.0 on several practical tasks:

• Creating infographics with data, coordinate axes, and captions
• Generating branded cards with text and logos
• Reproducing a specified design style — colors, fonts, composition
• Preparing visuals for social media and presentations
• Accuracy of text transmission — headings, bullets, numerical data

The overall conclusion — "impressive, but sometimes makes mistakes." Where requests are clear and layout is simple, results are consistently good. But on non-standard fonts, complex multi-column grids, or multiple text blocks simultaneously, the model can still produce artifacts and misalign elements.

Real-World Application

Despite the errors, Images 2.0 is already applicable to specific work tasks — and this wasn't obvious just a few months ago.

Quick illustrations for materials. For blogs, articles, and internal documents — generation takes seconds instead of hours in Figma or Photoshop. The result is not final design, but a good enough draft for first discussion.

Marketing visual prototypes. If you need to show a client or team an approximate appearance of a banner, card, or ad — Images 2.0 works as a tool for a first draft.

Infographics without a designer. With the right prompt, the model structures data and places it on an image — something that previously required Canva, Piktochart, or a separate task for a designer.

"This is more fun than I expected — and surprisingly useful for real

work," — ZDNet's testing conclusion.

Where Problems Still Exist

Images 2.0 works significantly better with specific and detailed instructions. Vague or overly general requests still often produce unsatisfactory results. Another downside — unpredictability: the same prompt in different requests can produce different results with varying quality levels.

There is no direct control over parameters: you cannot strictly define a grid, specify exact element placement, or use custom fonts. For professional design work, this remains a significant limitation.

What This Means

ChatGPT Images 2.0 is not a replacement for designers and not a "Midjourney killer." But it is the first AI generator in a mass-market product that really and consistently handles text and branding in images. For those who want to quickly create a working visual without special skills — this is already a working tool, not a demo effect.