OpenAI has introduced Deep Research, an advanced agentic AI system based on the latest GPT-03 model. Designed to handle complex knowledge work with high efficiency, Deep Research represents OpenAI’s ongoing efforts to push the boundaries of artificial intelligence in professional settings. This development comes as part of a broader trend of AI tools becoming integral to various industries.
In comparison to earlier AI models like DeepSeek R1, which previously challenged industry standards, Deep Research demonstrates significant improvements in reasoning and task performance. The rapid advancements highlight the competitive nature of AI development and the continuous pursuit of more capable and reliable AI solutions.
Enhanced Reasoning and Performance
According to OpenAI’s blog post, Deep Research outperforms DeepSeek R1 in logic-intensive tasks, achieving a 26 percent accuracy on Humanity’s Last Exam benchmark, nearly three times higher than R1’s 9 percent.
“Deep Research could do a single-digit percentage of all economically valuable tasks in the world,”
CEO Sam Altman stated, emphasizing the model’s potential in various applications.
Advanced Features and Capabilities
Deep Research offers functionalities such as real-time web scraping and the ability to compile comprehensive financial reports, scientific studies, and policy evaluations within minutes. Similar to DeepSeek R1’s “DeepThink” mode, it visually displays its reasoning process before delivering answers, enhancing user transparency and understanding of the AI’s decision-making.
Market Availability and Access
The AI model is available to ChatGPT Pro subscribers at a cost of $200 per month, with a limitation of 100 queries each month. Targeted primarily at professionals engaged in intensive knowledge work, Deep Research also supports general use cases like grocery shopping and car purchasing by generating detailed reports complete with charts and citations.
“These agents are coming for knowledge worker jobs, from researchers and tax preparers to customer service reps and beyond,”
noted Zach Lloyd, founder and CEO of Warp, an Altman-backed AI startup. Additionally,
“An A.I. model cannot be a censored propaganda machine for China or, even worse, propagating falsehood,”
Aravind Srinivas, co-founder and CEO of Perplexity AI, criticized DeepSeek R1’s answer bias, highlighting concerns over data reliability and censorship. While Deep Research shows notable progress, some experts like Grant Passmore of Imandra argue that the current accuracy is insufficient for real-world applications, highlighting the need for further integration of symbolic AI and automated reasoning.
Deep Research demonstrates OpenAI’s commitment to developing AI tools that enhance professional tasks by providing precise and reliable information swiftly. As the AI landscape evolves, such advancements are expected to significantly impact how industries operate and how professionals utilize AI in their daily workflows.
