From search to creation: how Perplexity is rewriting the AI rulebook
Perplexity launches Labs: it’s no longer just a search engine, but a full-fledged AI environment for building reports, web apps, and data automation.
On May 29, 2025, Perplexity made a bold announcement on X: it’s no longer just helping users find information — it’s helping them build. With the launch of its new “Labs” mode, the platform is transforming from an AI search engine into a full-fledged development environment for reports, dashboards, web apps, and automation scripts. This isn’t just an update — it’s a strategic leap into a new market.
More than chat or search
In the race for user attention, Perplexity is betting not only on accuracy, but on capability. Labs is a mode that allows users to write and execute code, create interactive interfaces, generate visualizations, and process large amounts of data — all without needing to be a developer or data analyst.
The key idea here is removing the barrier between ideas and execution. What used to require a team of developers, analysts, and designers is now accessible to a single user with a $20 subscription.
What does Labs actually do?
Labs introduces four core capabilities:
- Code generation and execution. Users enter a request — Perplexity writes and runs code to process data, generate documents, or automate tasks.
- Visualizations. Charts, graphs, and other visual elements can be created with just a few clicks to make data clear and digestible.
- Interactive web apps. Users can build simple web interfaces right within the platform — no third-party services required.
- File organization. All created files are stored in a dedicated “Assets” tab for quick access and reuse.
Built for deeper tasks
Unlike typical search or Research mode queries, Labs is designed for tasks that require up to 10 minutes of autonomous AI work. It combines deep search, code execution, and data visualization — effectively turning the browser into a lightweight, intelligent development suite.
These capabilities can be game-changing for entrepreneurs, analysts, marketers, or educators who need to automate reporting, build a prototype web app, or prepare a complex presentation — all without a technical team.
Read more DeepSeek’s New AI Model Update
A direct answer to the competition?
Absolutely. Sounds ambitious, doesn’t it? But how exactly can Perplexity Labs change users’ daily routines, and how does its approach differ from the tools we’re already used to?
Imagine a marketer who needs to quickly analyze the effectiveness of their latest advertising campaign. Instead of spending hours collecting data from various sources and manually building charts in Excel, they can simply give Labs a command: “Analyze data from Google Analytics and Facebook Ads for the last month, visualize the dynamics of clicks and conversions, and create a brief report.”
Perplexity won’t just find the information but will also perform the analytical work, providing a finished result. Or consider a small startup that needs a simple landing page for a new product. Instead of searching for a freelancer or learning website builders, the founder can describe the desired outcome, and Labs will generate a basic web application. “It’s like having a personal junior developer and analyst available 24/7,” shares one of the platform’s beta testers.
Against the backdrop of giants like ChatGPT, Gemini, and Claude, which mostly focus on generating text, code, or answers to queries, Perplexity Labs is betting on a final, actionable result. If the main models provide you with “bricks” (text, code), Perplexity aims to immediately build a “house” (a report, a web application, an automated process). This shifts the emphasis from simply receiving information to actively applying it and creating new value directly within the interface.
This step could be key in the battle for users who are increasingly looking not just for answers, but for ready-made, turnkey solutions. And although there is still much work to be done in improving and expanding functionality, Perplexity’s claim to leadership in the niche of “AI assistants for real tasks” looks very convincing.