Powered by Smartsupp

Claude’s Memory Gets a Major Upgrade: Anthropic Merges Systems to End Repetitive AI Briefings



By admin | Aug 25, 2026 | 5 min read


Claude’s Memory Gets a Major Upgrade: Anthropic Merges Systems to End Repetitive AI Briefings

An update to Claude’s memory system is set to remove one of the most irritating hurdles of using AI agents—the constant need to re-explain information the model already knows. On Tuesday, Anthropic revealed it is unifying the memory functions of its chat interface and Claude Cowork, meaning Claude will retain knowledge gained in one setting and apply it in the other. Additionally, Anthropic is making Claude’s stored memories visible, giving users the ability to view, modify, or delete any saved details.

Image Credits:Anthropic

This change is designed to create a smoother experience for users, allowing them to transition effortlessly from research to execution. Previously, you might have built up project ideas over time in conversations with Claude, but when it came time to put those plans into action, you’d have to start from scratch by feeding Cowork every piece of context. That process was a major source of frustration. Now, Claude operates as a single, cohesive assistant rather than two separate tools under one brand. Anthropic illustrates this with an example: if you need to draft a status update for your manager or prepare an agenda for an upcoming conference, Cowork would already grasp the relevant details from your past chats. It would know the expected headcount, the city, and the list of speakers, eliminating the need to re-enter that information.

Image Credits:Anthropic

Moreover, Claude will now add topics to its memory in real time as you converse, rather than waiting to summarize a session after it ends. This makes it much simpler to switch back and forth between chat and Cowork, since new memories are updated instantly and shared across both experiences, even mid-conversation. By default, Anthropic states that Claude will not store personal or sensitive data such as health information, race, ethnicity, religious beliefs, political views, gender identity, and similar categories. However, users can choose to include these by switching on the “include sensitive topics in memory” option. The app will also alert users whenever a sensitive topic is added to memory. That said, the company notes that Claude will never save certain items, including government-issued IDs, Social Security numbers, criminal records, immigration status, or anything else that conflicts with its acceptable use policy. This memory feature is automatically enabled for Free, Pro, and Max plans across web, desktop, and mobile platforms. For iOS and Android users, updating to the latest app version is required to access the feature.




RELATED AI TOOLS CATEGORIES AND TAGS

Comments

Please log in to leave a comment.

KevinPet 57 minutes ago

Автошкола Р-авто https://автошкола.рф/entity Рейтинг: 5 из 5 - 1 отзыв https://автошкола.рф/ Автомобили принадлежат школе, а не инструкторам https://автошкола.рф/teacherprepare Специалисты ДОСААФ следят за регулярностью технического обслуживания, страховкой и чистотой https://автошкола.рф/specialsignals Чтобы практический курс проходил безопасно и комфортно https://автошкола.рф/mytishchi Во-вторых, у вас будет возможность попрактиковаться в управлении транспортным средством на специальном полигоне и в городских условиях, чтобы отточить свои навыки https://автошкола.рф/teacherprepare 39 500 руб https://автошкола.рф/teacherprepare по новым программам https://автошкола.рф/korolev