ChatGPT unveils Deep Research: An AI research assistant

OpenAI has introduced a groundbreaking feature for ChatGPT called “deep research,” an AI agent capable of autonomously conducting in-depth research on the internet. This new capability allows ChatGPT to plan, execute, and refine research strategies, effectively mimicking the actions of a human researcher.

Read: Boom Supersonic XB-1 breaks the sound barrier, a major milestone for supersonic flight

Deep research leverages OpenAI’s latest models to delve into complex topics, analyze information from various sources, and synthesize findings into comprehensive reports. Users can interact with deep research using text, images, and even upload files like PDFs and spreadsheets. The AI then independently explores the web, gathers relevant data, and presents the findings with citations and a summary of its research process.

While still under development, deep research demonstrates impressive capabilities. In internal testing, it achieved a 26.6% accuracy on the “Humanity’s Last Exam” benchmark, surpassing GPT-4o’s 3.3% and significantly outperforming other AI models. This benchmark assesses the ability of AI models to answer expert-level questions.

Deep research is currently available to ChatGPT Pro users with a limit of 100 queries per month. OpenAI plans to expand access to Plus, Team, and Enterprise users in the future.

This significant advancement positions OpenAI at the forefront of AI research, demonstrating the potential for AI agents to revolutionize how we access and utilize information. As AI technology continues to evolve, we can expect to see even more sophisticated and impactful applications of AI agents in the near future.