ImageGen 2.0 is impressive but not what businesses asked for

OpenAI launched ImageGen 2.0 in ChatGPT on 21 April 2026. It renders text accurately, produces 2K visuals, and edits images through conversation. Social media is already flooded with examples. It is available on every ChatGPT plan, including Business and Enterprise.

Availability is not the same as relevance.

What ImageGen 2.0 does

OpenAI's new image model is a genuine step forward. It produces visuals at up to 2K resolution across multiple aspect ratios. Text rendering, the traditional weakness of AI image generators, now works reliably enough to create restaurant menus, social media graphics, and marketing layouts.

A paid-only Thinking mode adds reasoning, multi-output generation, and access to tools like web search. Users can refine images through conversation. Zoom in, adjust elements, change compositions, all without starting again. The model holds context across edits.

TechCrunch tested it with a Mexican restaurant menu. The result was immediately usable. Two years ago, the same request produced misspelled items like churiros and burrto. That progress is real.

The Studio Ghibli pattern

This is not the first time OpenAI has shipped a feature designed to go viral.

In early 2025, Studio Ghibli-style image generation dominated social media for weeks. Millions of users uploaded selfies and shared the results. Sign-ups spiked. Headlines followed. It was a masterclass in consumer marketing.

It produced zero measurable business value.

ImageGen 2.0 follows the same playbook. The launch is timed for maximum visibility. The demos are designed to impress on social feeds. The conversation is about what the model can do, not what your business needs it to do.

Consumer virality is cheaper than enterprise sales teams. Every viral moment drives sign-ups, media coverage, and brand awareness. OpenAI is building its user base from the bottom up.

The question for business leaders is whether you are adopting tools because they solve your problems, or because they made an impression on social media over the weekend.

What businesses actually need from AI

When gecco works with UK SMEs on AI adoption, image generation rarely features in the top 10 priorities.

Teams ask for help with document processing, customer response drafting, internal knowledge retrieval, and workflow automation. They want to reduce the 45 minutes spent writing a proposal or the two hours lost chasing information across three systems.

Those are measurable problems with measurable returns.

AI-generated images can play a supporting role. A small marketing team producing social content could save time on stock image searches or quick visual assets. That is a valid use case. But it is a narrow one, and it requires governance that most SMEs have not built yet.

Who approves AI-generated visuals before they go out? What happens when a generated image contains inaccurate text, a wrong logo, or a misleading claim? Who checks that the output meets your brand standards?

At gecco, we frame AI adoption as 80% people and culture, 20% technology. ImageGen 2.0 is impressive technology. But without clear policies on when and how your team uses it, you are adding risk alongside capability.

How to judge AI features that land on your desk

New AI capabilities arrive weekly. Not all of them deserve your attention. Before investing time in any new feature, run it through three questions.

Does it solve a problem your team already has? If nobody in your business has asked for AI-generated images, this feature does not become a priority because it trended on LinkedIn.

Can you govern it? If you have no policy on AI-generated visual content, publishing AI-made marketing assets is a brand and compliance risk. Build the governance before you adopt the tool.

Does it save measurable time or money? A feature that produces a wow demo but no repeatable workflow is entertainment, not productivity. Track the before and after.

If the answer to any of those is no, the feature is a distraction. Park it and revisit when the answers change.

Focus your AI investment on what delivers results

OpenAI will keep shipping features designed to go viral. That is their growth model, and it works. But your AI strategy should not be driven by someone else's marketing calendar.

Start with the problems your team faces every day. Build capability around those. Measure the results.

gecco helps UK SMEs adopt AI in a way that delivers measurable value, not hype cycles. If you want a clear picture of where your business stands, take the AI readiness assessment.