Perplexity has introduced a significant upgrade to its Comet Assistant: built-in memory that automatically stores user preferences, interests, and past conversations.
Instead of relying solely on long context windows or manual summaries, the system now preloads essential context, allowing the assistant to deliver more accurate and personalized responses without disrupting the workflow.
This allows Perplexity to act more like a true second brain, maintaining continuity across tasks and sessions.
Unlike traditional AI chatbots, which treat user history as training data, Perplexity retrieves specific memories from a user’s personalized store and uses them directly to generate more relevant answers.
This enables refined personalization, such as recommending running shoes based on injuries or training goals, recalling prior advice, or tailoring book suggestions to personal taste.
As users interact more, their memory stores grow more nuanced, improving precision over time.
Perplexity also emphasizes user control. Memory can be turned off at any time, is automatically disabled in incognito mode, and is fully encrypted.
Users can choose any model while maintaining the same personalized context across them, ensuring continuity even as they switch between faster, reasoning-focused, or domain-specific models.
The system is designed to preserve the value of users’ accumulated preferences and context, making every new model instantly more useful.
Key Takeaways:
- Perplexity introduced memory-powered personalization across its Comet Assistant.
- Users retain full control with encryption, toggles, and incognito mode protections.
- Memory persists across all models, allowing context portability and more tailored results.
You may also want to check out some of our other recent updates.
Wanna know what’s trending online every day? Subscribe to Vavoza Insider to access the latest business and marketing insights, news, and trends daily with unmatched speed and conciseness! 🗞️
