OpenAI Fights Back – NEW DEEP AGENTS!

OpenAI has launched a new model called Deep Research, which excels in conducting complex research tasks autonomously and has achieved a score of 26.6% on the benchmark Humanity’s Last Exam, significantly outperforming its closest competitor. This model enhances ChatGPT’s capabilities by synthesizing information from various sources and providing comprehensive reports, making it a valuable tool for professionals and content creators in need of accurate and detailed research.

OpenAI has introduced a groundbreaking new model called Deep Research, which represents a significant advancement in AI research agents. This model has achieved an impressive score of 26.6% on a challenging benchmark known as Humanity’s Last Exam, more than doubling the score of its closest competitor, Deep Seek R1, which scored only 9.4%. The Deep Research model is designed to autonomously conduct complex research tasks for up to 30 minutes, making it a powerful tool for projects requiring detailed information, such as writing a book, exploring business ideas, or making significant purchases.

The Deep Research model enhances the capabilities of ChatGPT by allowing it to perform multi-step research tasks and synthesize information from various online sources. Users can provide prompts, and the AI will ask clarifying questions to ensure alignment with the project objectives. The model then conducts research, tracking its tasks in a sidebar, and ultimately outputs a comprehensive report that includes relevant data, references, and recommendations. This functionality positions the AI as a valuable research assistant, capable of delivering insights that would typically take hours to gather.

The video highlights several practical applications for the Deep Research model, particularly for professionals engaged in knowledge-intensive work, such as finance, science, policy, and engineering. It is also beneficial for writers, journalists, and content creators who require reliable research. The model’s ability to synthesize large amounts of information quickly makes it an essential tool for anyone needing accurate and detailed research, whether for personal or professional purposes.

Examples showcased in the video illustrate the model’s capabilities, such as compiling data on mobile adoption rates and providing market recommendations for a new translation app. The Deep Research model not only presents information in a clear format, such as tables, but also cites specific sources, reducing the risk of hallucinations—where the AI generates incorrect or fabricated information. This feature is particularly valuable for entrepreneurs and business owners seeking to validate ideas and explore market opportunities.

OpenAI plans to roll out the Deep Research model to Pro users initially, with a limited number of queries per month, followed by broader access for Plus and Team users. The company envisions a future where AI can autonomously conduct research and perform related tasks, paving the way for more sophisticated applications of AI technology. As the capabilities of AI continue to evolve, the integration of research and operational agents could lead to even more advanced functionalities, enhancing the role of AI in various aspects of life and work.