Anthropic adds Memory to Claude for Team and Enterprise plan users

Anthropic has added a new Memory feature to its generative AI-based chatbot Claude for Team and Enterprise plan users to boost productivity.

The new optional feature will allow admins and users to control what Claude remembers about their chats, otherwise known as context.

Retaining context in generative AI-based tools like Claude, ChatGPT, and Microsoft Copilot helps users improve productivity as they don’t have to redefine concepts or provide background information while inputting a prompt or a query every single time.

In an enterprise setting, context would mean Claude remembering a team’s processes, client needs, project details, sales preferences, and other priorities as set by the user group or administrator.

Further, Anthropic has integrated the Memory feature to Claude Projects, its personalized workspace offering, available as part of the paid plans.

The integration will allow enterprises to separate memories of different projects so that Claude retains distinct memories of each of them in their individual workspaces, and these memories remain distinctly different from Claude’s context in general operations if it is deployed throughout an enterprise.

“These project boundaries (memories) help you and your teams manage complex, concurrent initiatives without mixing unrelated details, serving as a safety guardrail that keeps sensitive conversations contained,” the company wrote in a blog post.

However, a Memory feature in a generative AI-driven chatbot or assistant is not new, and rivals such as OpenAI and Google have introduced similar features in their respective offerings.

However, Forrester VP and principal analyst Charlie Dai pointed out that Claude’s implementation of the feature stood out.

“Instead of a global persistent profile, Claude uses project-scoped memory with strict isolation, and it requires explicit activation. This design prioritizes privacy, prevents context leakage between clients or projects, and offers granular control,” Dai said.

“While slower to market compared to rivals, this approach aligns with Anthropic’s safety-first philosophy and enterprise compliance needs,” Dai added.

Editing memories, incognito chat, and importing context

Anthropic is also providing administrators with the option to edit memories, which are stored in a memory summary inside Claude’s settings.

Administrators, according to the company, can update the summary at any time by chatting with Claude, and it will either focus on or ignore topics based on the new inputs.

Further, Anthropic has added an incognito chat mode — an option where users can talk to Claude without using or adding to the memory it has.

“It is perfect for sensitive brainstorming, confidential strategy discussions, or when you simply want a fresh conversation without context from previous chats,” the company wrote.

Subscribers of Team plans will also have the option to import and export memories from and to other AI services.

However, there are many caveats to successfully importing and deploying memories from other chatbots in Claude.

For one, Anthropic itself warns that memory imports are “experimental and still in active development, and at this stage, Claude may not always successfully incorporate imported memories.”

Secondly, users need to check if other chatbots allow access to memory or context.

Imported memories, according to Anthropic, may take 24 hours to appear as Claude processes memory updates only once daily.

“Importing a memory does not overwrite an existing memory,” the company said.

In the case of exporting memory from Claude, users can access the memory in settings and copy-paste it to another chatbot.

Forrester’s Dai pointed out that Anthropic’s move to allow importing of context from other chatbots could signal a push towards interoperability.

“This is an important step to reduce lock-in and foster competitive, multi-vendor AI ecosystems,” Dai said. Memory as a feature can be toggled on from Claude’s settings.