Archive — US — techcrunch
OpenAI launches new tools to help businesses build AI agents
OpenAI has launched new tools, including the Responses API, to help businesses develop AI agents for independent task completion. These tools aim to enhance AI autonomy and accuracy, replacing the older Assistants API. With rising hype around AI agents, OpenAI faces pressure to deliver functional and effective solutions as it attempts to mitigate common issues like hallucinations and unreliable citations in AI search.
Key points
- OpenAI released the Responses API for building AI agents to perform tasks like web searches and data retrieval.
- The Responses API replaces the older Assistants API, which will be phased out by mid-2026.
- OpenAI's AI models, such as GPT-4o search, claim high accuracy rates compared to previous models, with factually correct outputs but still prone to errors.
- New tools include capabilities for file searching and the Computer-Using Agent model for automating computer tasks.
- Despite improvements, existing challenges like AI hallucinations and unreliable citations remain.
- OpenAI offers an open-source toolkit, the Agents SDK, to assist developers in integrating these tools and ensuring reliability.
- OpenAI aims to shift from demo-based AI to practical applications, predicting a significant impact from AI agents by 2025.
Why it matters
- The rise of AI agents aligns with the increasing focus on automation and AI integration in the workforce.
- Similar trends are observed among tech companies exploring autonomous systems and intelligent agents.
- OpenAI’s initiative could influence the competitive landscape by establishing standards and functionalities for AI agent development.
- Concerns around AI transparency and reliability echo challenges from AI developments observed in recent years.
OpenAI's launch of the Responses API and associated tools represents a significant step toward refining AI agents, aiming to translate potential into practical applications amidst ongoing industry uncertainty.