Anthropic's AI chatbot Claude gets memory feature to recall past chats

Claude's new memory feature lets users resume chats and projects on demand, keeping context without starting over or storing details automatically

Anthropic has enabled memory feature on its Claude chatbot, making it easier for users to continue conversations and projects from where they left off. According to Anthropic, the AI chatbot can now reference past chats when asked, so users do not have to re-explain themselves. Additionally, Anthropic has posted a YouTube video showing how it works. 

 

In April, OpenAI introduced a major update to ChatGPT’s memory function, enabling its chatbot to recall previous conversations and offer more tailored responses.

 

In the demo, a user asked Claude what they had been working on before a vacation. Claude searched through their old chats, gave a list of topics and the summary, and asked if the user wanted to continue the same project. 

 

“Never lose track of your work again,” the company wrote. “Claude now remembers your past conversations, so you can seamlessly continue projects, reference previous discussions, and build on your ideas without starting from scratch every time.”

 

The memory feature works on web, desktop, and mobile versions of Claude. It can also keep different projects and workspaces separate, so you can find the right chat quickly. As of now, it is being rolled out to users with Max, Team, and Enterprise subscriptions. 

 

To turn the memory feature on:

• Go to Settings 
• Select Profile 
• Turn it on, Search and reference chats.

It is to be noted that this is not the same as ChatGPT’s “persistent memory.” Claude will not automatically remember everything you say or create a profile about you. Instead, it will only look up and reference your past conversations when asked.  

 

In recent news, Anthropic's Claude Code also features real-time security reviews, scanning AI-generated code for vulnerabilities like injection attacks and suggesting fixes to prevent flawed deployments. Developers using Claude Code can now activate an always-on mode where the AI monitors code changes, flags potential issues such as injection attacks or insecure data handling, and even suggests fixes.