AI apps are adding too many modes, making them harder to use

In short: Many popular AI apps are becoming harder to use because they split simple tasks across multiple named modes and features.

What's going on

Google recently promoted new voice features in its Gemini app and said people should not have to guess which tool to use for a task. But the Gemini app still asks users to pick between separate sections like Chat, Spark, and Daily Brief, each with its own label and icon.

TechCrunch argues this is confusing because users are being forced to learn the app’s “product architecture,” meaning how the company has organized its features behind the scenes (like being asked to choose the right department before you can ask a question at a store). The article points to Daily Brief as an example, which pulls info from Gmail and Calendar to send updates, but it may not know what is truly urgent versus what feels like an unwanted reminder. It can even resurface past searches, which some users may find uncomfortable.

The same pattern shows up in other AI apps. Anthropic’s Claude has “Chat” and “Cowork,” and ChatGPT has “Chat” and “Work,” which can make AI feel less natural to use. Instead of one simple place to ask for help, users have to decide which mode matches their goal.

What to watch

Some companies are trying the opposite approach by fitting AI into familiar places, not new modes. Apple has been adding AI improvements to existing tools like Siri and Spotlight Search, and some new services let people text an AI assistant like a contact (texting is a familiar interface, like sending a message to a friend). If this approach grows, expect more AI tools to hide their internal parts and focus on one clear way to ask for help.

Source: TechCrunch AI