AI BriefingOpenAI Blog news
The latest AI and developer news from OpenAI Blog, with short summaries.
Visit website →OpenAI Deploys Pando Git Backend, Cutting Median Push Latency by 80% and Handling 1PB Daily
OpenAI's Pando Git backend cuts median push latency by nearly 5x and delivers nearly a petabyte of Git data daily.
OpenAI Launches Grid Protectors Program to Support Reliable Electricity in Compute Campus Communities
The program commits to paying full infrastructure costs and reducing power draw during grid strain, with a pilot in Effingham County projected to save customers $950 million annually by 2029.
OpenAI Introduces LifeVal Benchmark to Measure AI Performance on Everyday Life Tasks
The evaluation, derived from time-use surveys and a study of 3,995 respondents, shows current models achieving scores between 90.4% and 100% on specific practical workflows.
OpenAI Appoints Dan Driscoll as Senior Counselor for National Security Modernization
Driscoll, the 26th Secretary of the Army, will advise OpenAI on applying advanced AI to strengthen national and cyber security for U.S. allies.
Qodo integrates organizational context into Codex to catch P0 bugs across 10,000+ repositories
The plugin provides Codex with codebase standards and cross-repository dependencies to identify critical issues before pull requests are opened.
OpenAI to Honor Common robots.txt Signal for AI Training Preferences
The move allows website operators to communicate content usage preferences to multiple AI providers at once, reducing the need to maintain separate rules for each crawler.
OpenAI Announces Community Safeguards Report and Expands Teen Protections Against Violence
OpenAI announced a new Community Safeguards Report and expanded restrictions on firearms content for minors, citing over 200 law enforcement referrals in Q3 2026.
Experts and US Government File Briefs Supporting Fair Use for AI Training in NYT Lawsuit
Amici curiae including the US Government, Authors Alliance, and independent creators argue that using copyrighted works for AI training is transformative and does not cause cognizable market harm.
OpenAI Habitat Cuts Read Latency from 120ms to 10ms with Rust Rewrite and CDC Caching
A Rust rewrite and CDC-based caching reduced global read latency by 12x.
OpenAI updates Model Spec to organize AI behavior around five core values
The revision retires detailed style guidance, adds new sections on context and oversight, and shifts terminology from 'the assistant' to 'GPT'.
OpenAI Expands Frontier Model Misalignment Monitoring to Cover Training and Deployment
The system covers training, evaluation, and deployment, with GPT-5.6 Sol showing superior detection rates compared to Terra and Luna variants.
Ropes & Gray develops proprietary AI tool for transactional diligence with OpenAI
The tool automates the first pass of reviewing thousands of pages, grouping documents by theme and generating reports in the firm's format for lawyer review.
Credit Saison Achieves 954% ROI in ChatGPT Enterprise Pilot, Driving Companywide Rollout
The pilot exceeded targets with 75% of managers reporting they could not do without the tool, leading to a companywide rollout one month after completion.
Evernote reports 10-20% higher trial-to-paid conversion from ChatGPT Ads
Campaigns focused on a single feature drove 25% of desktop visits to new accounts, with trials starting August 28-September 12, 2026.
OpenAI and ServiceNow expand partnership to integrate models into enterprise workflows
The expansion brings OpenAI models into ServiceNow products and connects ServiceNow workflows to ChatGPT and Codex, with the MCP plugin available now and various new capabilities expected between late 2026 and 2027.
OpenAI Researchers Propose Three Scenarios for AI Integration in Corporate Structures
Hemanth Asirvatham and Elliott Mokski argue that AI will function as active labor rather than passive tools, leading to hybrid firm structures where human value shifts to coordination and judgment.
Legora's agentic legal platform cuts NDA review time by 95% and boosts Fortune 100 client efficiency
Legora, a Stockholm-based legal-tech firm, uses OpenAI models to automate legal workflows, helping a Fortune 100 customer save over $20 million and reclaim 9.5 hours per lawyer per week.
PickOpenAI and Oracle integrate applications into ChatGPT and launch AI-powered products
Oracle reports 130,000 active ChatGPT users and 95,000 Codex users while launching Oracle AI Workspace and new security and SMB offerings powered by OpenAI models.
Check Point Uses OpenAI Daybreak to Cut Vulnerability Research Time from Hours to Minutes
The company leverages Daybreak's frontier cyber models to automate analysis and accelerate the development of security signatures for over 100,000 organizations.
PickGreptile Uses GPT-6 Astra to Automate Application Testing and Security
The startup leverages the model's computer use capabilities to simulate user interactions and detect UI bugs that static code review misses.