OpenAI, a major US tech company specializing in artificial intelligence (AI), has launched a ‘deep research’ feature for its ChatGPT users. The main aim of the company behind launching this AI agent is to support its users by providing the answers to complex research in ten minutes, which a human would take many hours to do.
This AI feature is developed to handle complex and difficult queries by providing proper and reliable answers to them. The AI feature is powered by a version of the upcoming OpenAI O3 model.
This tool is built for the use of professionals who are in fields such as finance, science, policy, and engineering, and for professional researchers who do thorough, precise, and reliable research.
It also aims to benefit individuals by analyzing proper research for making a careful decision about purchasing cars, appliances, or furniture.
As per the official blog, the tool is trained on real-world tasks with the use of a browser and Python tool and by using the same learning methods that were used for training the company’s OpenAI o1 model.
Currently, the company is also looking to expand the tool’s capabilities, as its outputs are limited to text; it plans to add images, data visualizations, and other analytical elements to its outputs.
OpenAI highlighted that its deep research process may take five to 30 minutes; additionally, its users will receive a notification once the task is completed.
The new model has shown excellent results on a number of public evaluations focused on real-world problems, including Humanity’s Last Exam. While other models struggle below 10%, Deep Research has a score of 26.6%.
OpenAI’s deep research is made for its users who need extensive research across multiple websites and sources, which is more than a simple summary.
The company has announced that the feature will be available to ChatGPT Pro users, stating that its pro users will be able to make up to 100 queries per month.
It further added that “more options for its Plus and its Team users are expected soon. To get access to its deep research, the user simply needs to select the option in the composer and enter their query. The tool can process inputs such as files and spreadsheets, which are useful for those seeking answers from a combination of data.”
However, currently, this feature is only available as a web experience, but still, the combination of mobile and desktop apps is expected later this month.
