OpenAI joined the agentic artificial intelligence (AI) race Tuesday, launching new tools for developers to build agents in a move that pits it not just against a controversial new Chinese AI startup but major investor Microsoft Corp.
OpenAI’s new Responses API (application programming interface) allows businesses to create customized AI agents capable of performing web searches, navigating websites and scanning company files, as its Operator model does.
“The Responses API is our new API primitive for leveraging OpenAI’s built-in tools to build agents. It combines the simplicity of Chat Completions with the tool-use capabilities of the Assistants API. As model capabilities continue to evolve, we believe the Responses API will provide a more flexible foundation for developers building agentic applications,” OpenAI said in a blog post.
OpenAI’s new tools represent a major improvement in autonomy from what it currently offers. In January, the Silicon Valley startup debuted two AI agents in ChatGPT — Operator, which navigates websites, and deep search, which compiles research reports. More recently, CEO Sam Altman shared on X, “we trained a new model that is good at creative writing (not sure yet how/when it will get released). this is the first time I have been really struck by something written by AI; it got the vibe of metafiction so right. PROMPT: Please write a metafictional literary short story,” teasing newer upcoming models.
On Tuesday, OpenAI also said it inked a five-year, $11.9 billion deal for cloud-infrastructure startup CoreWeave to provide AI infrastructure services ahead of its much-anticipated initial public offering (IPO) this year. Under the deal, OpenAI will receive a $350 million stock stake in CoreWeave, whose IPO is expected to fetch it a market value of up to $35 billion.
OpenAI’s flurry of AI moves come days after a report in The Information that Microsoft, which has injected $13 billion into OpenAI, is testing models from OpenAI, DeepSeek, Meta Platforms Inc. and Elon Musk’s xAI, as potential OpenAI replacements in Microsoft Copilot.
OpenAI is also eyeing competition from Chinese startup Butterfly Effect who earlier this week went viral with Manus, a new AI agent platform that claims to be “the first general AI agent” capable of performing complex, multi-step tasks such as screening resumés and creating a website. Some tech experts, however, said the model’s platform was built from Anthropic’s Claude and Alibaba’s Qwen to analyze financial filings and draft research reports, and that its performance is spotty.
OpenAI is but the latest company to jump into the agentic AI fray. In recent days, ServiceNow Inc., Salesforce Inc., Microsoft, Amazon.com Inc.’s AWS and Cisco Systems Inc. all presented their agentic AI plans with products and services in pursuit of a booming industry. ServiceNow also announced a $2.85 billion acquisition of AI assistant maker Moveworks on Monday. The AI agent market is expected to skyrocket to $40 billion in 2029 from $5 billion in 2024, says AI agent researcher Mark Stansberry.
“Software organizations that embrace agentic AI and AI-first development paradigms early will be best positioned to capitalize on its benefits, from increased developer productivity to faster software delivery,” Mitch Ashley, vice president and practice lead of DevOps and AppDev at The Futurum Group, said in a report issued this week.
“This new wave of agentic AI development tools represents a shift from AI-augmented tools, which supports developers in writing code, to agentic AI, which actively performs multi-step development tasks,” Ashley added.
