Aimed at Professionals, New AI Service Offers Speedy Research with Limitations

OpenAI has unveiled its latest innovation, the ‘Deep Research’ tool, in response to growing competition from Chinese AI companies, notably DeepSeek. Announced on February 2, the tool is designed to accelerate online research, enabling paying users of OpenAI’s ChatGPT chatbot to generate comprehensive research reports by analysing numerous sources within minutes.

Deep Research promises to reduce research time dramatically, claiming that tasks which usually take hours can now be completed in a matter of minutes with a single query. It is primarily targeted at professionals in sectors such as finance, science, engineering, and policy, and is intended to streamline and simplify complex, time-consuming research tasks. The tool’s high degree of accuracy and precision is meant to offer an alternative to traditional, manual research methods, saving valuable time for users.

This release marks OpenAI’s second major tool launch of 2025, following January’s introduction of Operator, a tool designed to assist with daily tasks like booking flights, ordering groceries, and making purchases. Like Operator, Deep Research is currently exclusive to ChatGPT Pro users, with a monthly subscription fee of US$200 (approximately S$273).

The introduction of Deep Research comes as OpenAI faces increasing competition from a range of global AI companies, including Microsoft and Anthropic, along with startups that are developing their own AI tools. DeepSeek, a Chinese AI firm, has made notable strides, prompting concerns among Western developers about the rising threat posed by these competitors.

Despite the tool’s impressive capabilities, OpenAI has acknowledged that Deep Research is still in its early stages and has some limitations. While it produces more accurate responses than previous iterations of ChatGPT, the tool can occasionally “hallucinate” facts, make incorrect inferences, or confuse authoritative information with unverified rumours. Additionally, it may struggle to clearly express uncertainty or make accurate citations.

OpenAI assured users that these issues would be addressed over time through further refinement. Initially, users will be restricted to 100 queries per month, though the company plans to extend access to Plus, Team, and Enterprise subscribers in the future, with no set timeline for this expansion.

As the AI landscape becomes increasingly competitive, the launch of Deep Research represents a significant move by OpenAI to remain at the forefront of AI-driven research tools.