ZDNet compared ChatGPT and Claude: OpenAI stronger on features, Anthropic on writing and logic

On April 1, 2026, ZDNet published a practical comparison of ChatGPT and Claude and ran both chatbots through 10 identical scenarios. The result turned out inconvenient for fans of simple rankings: there is no single absolute leader, but the difference between the models became more noticeable precisely in everyday work.

Where Claude is Stronger

Based on the test results, Claude performed better in areas where AI is expected to provide not flashy demonstrations, but careful thinking. In text tasks, explanations, and research queries, Anthropic's model more often produced more natural language, better maintained tone, and less frequently fell into template formulations. This is especially noticeable in scenarios where you need to write a message, rework text, compile an answer from multiple conditions, or quickly extract meaning from a large volume of information.

A separate plus for Claude is more confident work with nuances. If a question doesn't have one obvious answer, the model more often shows its line of thinking, acknowledges limitations, and doesn't rush to make overly broad conclusions. In ZDNet's article, this is presented as the main argument in favor of Claude for people who use AI not for entertainment, but as a working tool for notes, analysis, and drafts.

In other words, Claude turned out to be closer to the role of editor and analyst than to the role of a universal digital assistant.

Where ChatGPT is Ahead

ChatGPT has a different zone of strength: multimodality and ready-made user functions. In the test, it looked more confident in tasks where not only a text answer mattered, but the entire surrounding experience — voice mode, image generation, integrations with external services, and more convenient access to additional tools. If Claude wins in depth and carefulness of response, ChatGPT more often wins in how quickly it turns the answer into action or a finished artifact.

• Voice scenarios and smoother conversational experience
• Image generation and editing directly in the product
• Integrations with external services and applications
• A broader set of features for everyday tasks

That's why ChatGPT in such a comparison looks stronger for the average user. When you need not just to get good text, but, for example, quickly assemble a visual, use voice, connect a third-party service, or go through a short everyday scenario without extra setup, OpenAI's product turns out to be more practical. This doesn't make it "smarter" in the general sense, but it makes it more convenient in situations where speed, format, and minimum friction are important.

The Cost of Transition

One of the most useful parts of the comparison is the conclusion that a complete move from one chatbot to another is rarely rational. ZDNet separately notes the limitations of free tiers: both platforms fairly quickly hit rate limits, and users may face downgrade to less powerful modes. Because of this, the impression of the service depends not only on the model, but also on what time of day you use it, how long the session lasts, and whether you run into limitations on messages or tools.

From this follows the main practical conclusion of the article. If your work is tied to texts, reworking materials, structuring thoughts, and more cautious answers, Claude looks stronger. If you need a wide combine with voice, pictures, and connections with other services, it makes more sense to stay with ChatGPT.

The test itself essentially destroys the idea that one unconditional winner has already appeared on the market. Quite the opposite: the more actively AI is embedded in everyday processes, the more important it becomes not the brand, but the exact correspondence of the tool to the task.

What This Means

ZDNet's comparison shows that the AI assistant market is rapidly moving away from the "choose one service and live with it" model. In 2026, a more practical strategy is to keep several tools nearby and distribute roles: Claude for writing and reasoning, ChatGPT for multimodality, voice, and integrations.