cd /news/artificial-intelligence/claude-can-now-tell-you-that-you-use… · home topics artificial-intelligence article
[ARTICLE · art-54066] src=thenextweb.com ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

Claude can now tell you that you use Claude too much

Anthropic launched Reflect, a beta dashboard for its Claude chatbot that tracks usage patterns and suggests breaks, aiming to promote thoughtful AI use rather than maximizing screen time. The feature, developed with MIT Media Lab and other partners, offers quiet hours, break reminders, and prompts to reflect on personal tasks, though critics argue it may deepen user dependency on AI.

read4 min views1 publishedJul 10, 2026
Claude can now tell you that you use Claude too much
Image: Thenextweb (auto-discovered)

Anthropic has built a feature that does something odd for a tech company. It counts how much you lean on its product, then gently suggests you might want to lean a little less.

The feature is called Reflect, and it landed in beta on Thursday. It is a dashboard, tucked into Claude’s settings, that shows you how you have been using the chatbot. Anthropic describes it as a way to “see your patterns and shape them”. Most of the press has a blunter name for it: Claude Wrapped.

The comparison is fair. Like Spotify’s yearly recap, Reflect serves up a tidy summary of your habits. It lists your key topics, your busiest day, your peak hour, and your total number of chats. You can look back over one, three, six, or twelve months.

The Screen Time twist #

Where it departs from Wrapped is the mood. This is less a celebration than a nudge. Reflect borrows heavily from the digital-wellbeing playbook, closer to Apple’s Screen Time than to a party.

You can set quiet hours. You can schedule a reminder to take a break after a while. Both are dismissible, so they never block you mid-task. Every so often, the dashboard also poses a pointed question, such as:

“What’s one thing you want to keep doing yourself, even if Claude could do it faster?”

Then comes the part The Verge could not resist flagging. Once you answer that question, Reflect offers to talk it through with you. So the tool built to help you use Claude less invites you to discuss the matter with Claude.

A coach for how you prompt #

The dashboard is not only about cutting back. It also tries to make your usage sharper. It sorts your habits into a four-part scheme Anthropic calls the AI Fluency Framework: delegation, description, discernment, and diligence.

From that, it offers concrete tips. If you keep re-explaining the same background, it may suggest starting a Project so you stop repeating yourself. In one reviewer’s case, it proposed building a custom fact-checking skill, then wrote a template that made Claude cite its sources and flag its confidence. On privacy, Anthropic has drawn some clear lines. Reflect ignores incognito chats. It does not pull in the underlying files from your connected tools, so a summarised inbox might show up while the emails themselves do not.

Anything tied to a health integration is left out completely. Sensitive topics can surface, but only at a high level.

Why a chatbot maker wants you to log off #

The obvious question is why an AI company would build a brake pedal for its own product. Anthropic’s answer is that better use beats more use.

“We were really intentional about building it with an eye toward how we can upskill people’s usage of Claude, not in a way that encourages them to spend more time with it,” Ryn Linthicum, the company’s head of wellbeing policy, told Engadget. Tellingly, the dashboard does not yet show total time spent.

Linthicum said that was a number the product team “didn’t want to maximise”, though it is coming later.

The timing is not an accident either. AI backlash is loud right now, from data-centre protests to a steady drip of studies warning about cognitive off. A tool that says “use me thoughtfully” is a neat answer to that mood.

Anthropic developed it with the MIT Media Lab, the Digital Wellness Lab at Boston Children’s Hospital, and the Family Online Safety Institute.

Or is it just good retention? #

Not everyone reads it so warmly. TechCrunch argued that Reflect is “quietly selling you on AI”. Laying out everything Claude has done for you, the argument goes, makes the tool feel indispensable. Nudging you towards Projects deepens the integration and makes switching to a rival harder.

There is history here. TechCrunch drew a line back to Gmail Meter in 2012, a stats tool that, while fun, also quietly showed how central Gmail had become to daily life. A wellbeing feature and a retention feature can be the same feature.

Both readings can be true at once. Reflect is a genuine attempt to make AI use more deliberate, and it keeps you inside Claude while you deliberate. For now it is in beta for Free, Pro, and Max users with memory switched on.

Support for Cowork and mobile is on the way. Whether it makes you close the tab, or simply open it more mindfully, may depend on which of those two purposes wins out.

Get the TNW newsletter #

Get the most important tech news in your inbox each week.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @anthropic 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/claude-can-now-tell-…] indexed:0 read:4min 2026-07-10 ·