OpenAI API

Evidence (16)

• 4The New Stack·10/8/2026research
  Designing Scalable APIs for AI Agents

  The article explains why AI agent traffic behaves differently from normal web traffic and argues that microservices-style design is better suited to handling it. It uses an OpenAI-compatible API example and a prototype backed by Oracle AI Database Free to show how shared state, retries, and bursty tool calls affect scalability.

• 4OpenAI Blog·10/7/2026product_launch
  Radisson adds hotel search and booking to ChatGPT

  Radisson Hotel Group partnered with Accenture to create a ChatGPT plugin that helps travelers discover, compare, and book hotels while planning trips. The feature uses OpenAI technology to bring hotel discovery into the chat experience.

• 7NVIDIA GenAI·10/1/2026model_release
  OpenAI GPT-6 Astra Ultrafast runs on NVIDIA Blackwell GPUs

  OpenAI says GPT-6 Astra Ultrafast is now available in the OpenAI API and for eligible ChatGPT Work and Codex users, with inference optimized for NVIDIA Blackwell GPUs. OpenAI claims the mode can generate tokens up to 8x faster than Astra Standard, aiming to improve coding-agent workflows and other latency-sensitive tasks.

• 7Hacker News·10/1/2026product_launch
  Pi 1.0 and Pi Durable launch

  Earendil has released Pi 1.0, a hardened minimal agent harness with new support for MCP, extension-based virtual models, deferred tool loading, and other usability improvements. The company also introduced Pi Durable, an experimental package for building longer-running agentic applications, both released under the MIT license.

• 5OpenAI Blog·10/1/2026product_launch
  Albertsons uses ChatGPT Enterprise to speed up grocery operations

  Albertsons Companies is adopting ChatGPT Enterprise and the OpenAI API to help its teams work more efficiently across the business. The company says the tools are also helping it improve the grocery shopping experience for millions of customers.

• 7The New Stack·9/29/2026product_launch
  OpenAI launches Decision API built on Luna

  OpenAI announced a new Decisions API at DevDay, positioning it as a fast, low-cost model for returning predefined answers with confidence scores. The API is in limited preview and appears aimed at classification, routing, and agent action selection, though pricing and other details are not yet public.

• 9InfoQ·9/29/2026product_launch
  OpenAI DevDay 2026 recap

  OpenAI recapped more than 20 announcements from DevDay 2026, spanning new model, product, API, security, and developer-tool updates. The event highlighted GPT-6 Astra, ChatGPT, Codex, and additional builder-focused capabilities across the OpenAI platform.

• 9Hacker News·9/18/2026regulation
  Microsoft and OpenAI face new copyright claims over AI training data

  Unsealed court filings in The New York Times’ lawsuit allege Microsoft and OpenAI privately acknowledged AI training data collection amounted to theft and could harm publishers’ business models. The filings also cite claims of paywall bypassing, mass scraping, copyright notice removal, and product behavior that may substitute for original news sources.

• 7The New Stack·9/11/2026framework_update
  OpenAI’s safety system is interrupting API responses mid-task

  OpenAI is considering slowing development of its most advanced AI systems as industry leaders debate the pace of frontier model release. The company has already restricted access to some high-risk capabilities under its Preparedness Framework, and some API users have seen model responses cut off mid-task when safety controls intervened.

• 9Ars Technica AI·9/9/2026security
  U.S. Accuses Six Chinese AI Firms of Copying Frontier Models

  U.S. agencies say DeepSeek, Moonshot AI, Alibaba, MiniMax, StepFun, and Z.AI have been conducting industrial-scale distillation attacks against frontier AI models from U.S. companies since at least late 2024. The agencies allege the firms likely operated with Chinese government awareness, citing use of capabilities extracted from models such as Claude, GPT, Gemini, and Grok.

• 7The New Stack·9/7/2026research
  OpenAI says AI agents are adding supervisory work

  OpenAI reports that its researchers are now using an "automated research intern" agent for multi-day, well-defined tasks, but the company says rising agent use has also increased the amount of human supervision required. By mid-August, agents were logging 3.1 agent-workdays for every human workday across the research organization, while researchers also faced significant inference costs and ongoing oversight demands.

• 8OpenAI Blog·9/1/2026security
  OpenAI says Astra reaches critical cybersecurity threshold

  OpenAI says Astra is the first model to meet its Critical cybersecurity capability threshold under the Preparedness Framework. The company says it will apply stronger safeguards before releasing the model.

• 4OpenAI Blog·8/31/2026product_launch
  Polimill uses OpenAI models to power municipal knowledge search

  Polimill is using OpenAI GPT models and Codex to help municipalities search and apply administrative knowledge more efficiently. The company says the tools are also speeding up development of its public-sector AI infrastructure in Japan.

• 8The New Stack·8/26/2026research
  OpenAI’s Astra Tests Powerful Persistent Coding Agents

  OpenAI’s unreleased foundation model, Astra, is already being used internally to turn research ideas into code, run experiments, and return results with minimal human intervention. The model may also have reached OpenAI’s highest cybersecurity threshold, highlighting the challenge of giving long-running agents more autonomy while keeping them contained and secure.

• 7OpenAI Blog·8/19/2026framework_update
  OpenAI Offers Zero Data Retention for Eligible API Customers

  OpenAI says eligible API customers can use Zero Data Retention, meaning their API inputs and outputs are not stored. The company also previewed Private Safety Processing, a privacy-preserving approach for advanced safety checks on sensitive data.

• 8OpenAI Blog·8/13/2026product_launch
  OpenAI and Cerebras Launch Ultrafast Mode for GPT-5.6 Sol

  Cerebras and OpenAI announced Ultrafast Mode, a new OpenAI API tier powered by Cerebras for GPT-5.6 Sol. The service is designed to deliver very high output speed, claiming up to 750 tokens per second without quality loss, and is initially available to a limited set of customers. The companies say the mode improves latency for time-sensitive workloads such as incident response, security operations, and agentic workflows, and cite benchmark results showing substantially faster execution on knowledge-work tasks.