OpenAI is enhancing its AI-powered “deep research” feature with the ability to analyze codebases on GitHub.
On Thursday, OpenAI announced what it’s calling the first “connector” for ChatGPT deep research, the company’s tool that searches across the web and other sources to compile thorough research reports on a topic. Now, ChatGPT deep research can link to GitHub (in beta), allowing developers to ask questions about a codebase and engineering documents.
The connector will be available for ChatGPT Plus, Pro, and Team users over the next few days, with Enterprise and Edu support coming soon, according to an OpenAI spokesperson.
The GitHub connector for ChatGPT deep research arrives as AI companies look to make their AI-powered chatbots more useful by building ways to link them to outside platforms and services. Anthropic, for example, recently debuted Integrations, which gives apps a pipeline into its AI chatbot Claude.
OpenAI years ago offered a plugin capability for ChatGPT, but deprecated it in favor of custom chatbots called GPTs.
“I often hear that users find ChatGPT’s deep research agent so valuable that they want it to connect to their internal sources, in addition to the web,” OpenAI Head of Business Products Nate Gonzalez wrote in a blog post on LinkedIn. “[That’s why] today we’re introducing our first connector.”
In addition to answering questions about codebases, the new ChatGPT deep research GitHub connector lets ChatGPT users break down product specs into technical tasks and dependencies, summarize code structure and patterns, and understand how to implement new APIs using real code examples.
Techcrunch event
Berkeley, CA
|
June 5
There’s a risk that ChatGPT deep research hallucinates, of course — no AI model in existence doesn’t confidently make things up sometimes. But OpenAI is pitching the new capability as a potential time saver, not a replacement for experts.
An OpenAI spokesperson said ChatGPT will respect an organization’s settings so users only see GitHub content they’re already allowed to view and codebases that’ve been explicitly shared with ChatGPT.
OpenAI has been investing in its tooling for assistive coding, recently unveiling an open source coding tool for terminals called Codex CLI and upgrading the ChatGPT desktop app to read code in a handful of developer-focused coding apps. The company sees programming as a top use case for its models. Case in point, OpenAI has reportedly reached an agreement to buy AI-powered coding assistant Windsurf for $3 billion.
In other OpenAI news on Thursday, the company launched fine-tuning options for developers looking to customize its newer models for particular applications. Devs can now fine-tune OpenAI’s o4-mini “reasoning” model via a technique OpenAI calls reinforcement fine-tuning, which uses task-specific grading to improve the model’s performance. Fine-tuning has also rolled out for the company’s GPT-4.1 nano model.
Only verified organizations can fine-tune o4-mini, according to OpenAI. GPT-4.1 nano fine-tuning, meanwhile, is available for all paying developers.
OpenAI began gating certain models and developer features behind verification, which requires organizations to submit an ID and other identity documents, in April. The company claims that it’s necessary to prevent abuse.
