Briefly, Perplexity AI is an AI-powered online search assistant that focuses on in-depth search and provides reliable information with citations, while ChatGPT excels at engaging in conversation, responding to complex dialogue, and generating contextually accurate text and images.
‍

ChatGPT dominates the generative AI market and has inspired whitepapers, reports, and even memes. However, it is far from perfect and users have raised numerous concerns about the efficacy and reliability of the generated information.

Enter Perplexity AI, a new contender aiming to address these accuracy concerns with its own generative AI tool. But can Perplexity AI bridge the gap and dethrone the current leader?

Perplexity AI has seen a steady gain in popularity, with a 23.8% increase in website visits just within February 2024, as compared to ChatGPT’s 2.85%. As of January 2024, Perplexity has over 10 million active monthly users -  indicating a growth rate of 1,000x in a year.

What's the reason behind Perplexity AI's popularity?

‍

‍

Comparison: Perplexity AI vs ChatGPT

Both Perplexity AI and ChatGPT are large language models (LLMs) with distinct capabilities and features.

While Perplexity AI operates like an intelligent search engine, scanning the internet for answers and providing citations, ChatGPT excels in content generation, offering a conversational interface for coherent, human-like responses.

Here’s a brief comparison between them:

‍

‍

Perplexity AI

When Perplexity AI is asked a question, it doesn’t return a list of blue links for the user to go through. Instead, it summarizes the information in context to the user query and provides citations for the user to verify the source.

How does Perplexity AI make this possible?

At its core, Perplexity AI uses OpenAI’s GPT technology but has been trained as a conversational search engine. Perplexity AI’s founder and CEO, Aravind Srinivas, has described this AI chatbot assistant as the future of search.
‍

Notable features:

• Real-time information: Perplexity AI stands out for its capability to fetch and synthesize information in real-time from across the web, allowing users to receive the most current answers to their queries.
• Focus search option: This feature allows users to refine their searches by selecting specific sources such as academic publications, YouTube, Reddit, or Wolfram | Alpha for a more targeted result.
• Citations: By providing citations for all its results, it allows users to verify the credibility of the information, hence there is minimal chance of hallucination or false information.‍
• Related question prompts: Based on your query, Perplexity AI suggests related questions that help you dive deeper and get more insights into the topic.
  ‍

‍

ChatGPT

ChatGPT doesn’t need an introduction, given its popularity.

ChatGPT was introduced as an LLM that allows users to engage in human-like conversations and perform tasks such as content creation, data analysis, and more. Moreover, with appropriate ChatGPT plugins, one can automate their daily workflow tasks.

From September 2023, ChatGPT has been a multimodal AI model which means it can generate images in the same interface using DALL.E and users can even create GPTs as per their requirement.
‍

Notable features:

• Contextual understanding: Its ability to provide relevant and coherent responses enhances user interaction, making it a suitable tool for engaging in complex dialogue and ensuring a seamless conversational experience.
• Knowledge base integration: A user can integrate ChatGPT with an external knowledge base via a Plugin, GPT Builder, and API. This ensures that responses are based on the latest developments, data, and facts.‍
• Multimodal AI model: As a multimodal AI model, ChatGPT can process and generate information beyond text, including images and potentially audio or video. This expansion broadens its application, making it more versatile in user engagement and content creation.

‍

‍

So, how should a user choose between these tools? 

The answer lies in their differences.

The core model of Perplexity AI and ChatGPT is the same - GPT4. However, by understanding the tools’ capabilities and limitations, a user can select the one that best suits their requirements. Here are a few parameters:
‍

1. Information Accuracy

Perplexity AI: Draws responses from live web pages, choose it when you want to ensure updated and accurate information, with source citations. 

ChatGPT: Generates human-like, contextual responses without disclosing its information sources. For the source of information, you need to use the plugin for web browsing or prompt it to share information.

2. Quality of Responses

Perplexity AI: Capable of providing insightful answers with citations, allowing users to explore complex questions with confidence in the information's source. 

ChatGPT: Its evolution is more focused on creating dynamic engaging interactions and content generation capabilities. Since it depends more on training data and a knowledge base for responses, it can hallucinate data.
‍

3. User Interface 

Perplexity AI: The interface offers a more modern and sophisticated experience with features like “Collection,” where you can organize your conversation threads in one place. It also has privacy settings for enhanced data control and confidentiality.

ChatGPT: UI is quite simple and easy to navigate, however, all your conversations are arranged chronologically in one endless scroll. It is not easy to search a conversation thread created, say, 4 months ago in ChatGPT.

‍

Ultimately, the choice of tool depends on how it aligns with your personal and/or professional objectives.

When should you use ChatGPT:

If you value interactive, contextual dialogue with options for image generation, text generation, API integration, Plugins, or creating GPTs for specific tasks, then ChatGPT is your choice. 
‍
When to use Perplexity AI:

Perplexity AI will be a better choice if your needs are akin to having a smart research assistant capable of scanning, summarizing, and attributing information with accuracy and depth.

‍Which multimodal AI tool or LLM have you been using?

Let us know your thoughts.

‍