OpenAI announcements

Diagram-MMU: A Multi-Modal Benchmark for Scientific Diagrams

How RingCentral builds AI-native work from engineering to ops

From assistance to execution: How enterprises put AI to work

Stealing Reasoning Traces from Proprietary LLM APIs

What building an AI-native finance function taught me

OpenAI CFO Sarah Friar shares five lessons for building an AI-native finance function, from automated forecasting to stronger controls and AI ROI.

OpenAI’s letter to Governor Abbott on responsible AI infrastructure in Texas

OpenAI sent Governor Greg Abbott a letter outlining its commitment to responsible AI infrastructure in Texas. The letter supports reliable, transparent growth that benefits Texans.

Grammar Engineering Meets LLMs: Development of Cantonese and Irish ParGram Treebanks

Blast Radius

NiyamAI - An Intent-Bound AI Agent with Cryptographically Verifiable Guardrails using Zero-Knowledge Proofs

LMM Modality Transfer: A Pre-requisite for Autonomous GIS Agents

Working with the American Psychological Association on youth mental health and AI

OpenAI and the American Psychological Association advance evidence-based guidance, resources, and safeguards for responsible AI use and youth mental health.

How HSP GRUPPE builds AI capabilities for tax advisory

Discover how HSP GRUPPE uses ChatGPT Enterprise to boost productivity, improve work quality, and create more capacity for tax advisory and client service.

How we built a realtime system for responsive voice AI in six months

GPT-Live enables continuous voice interaction with AI, using a turnless speech model and low-latency architecture for faster, more natural conversations.

Romance-baiting scam: AI-assisted pig butchering workflows

OpenAI banned accounts that appeared to originate in Cambodia and used AI to translate and generate romance, investment scam conversations.

Cyber threat actors: AI-assisted intrusion research

OpenAI banned accounts potentially associated with publicly reported DPRK-affiliated threat actors using AI to research intrusion tooling, phishing, malware, and cryptocurrency targeting.

Task scam: AI-assisted fake review jobs

OpenAI banned accounts that appeared to originate in Cambodia and used AI to translate messages for fake-review job scams that asked victims to pay fees.

Operation “Peer Review”: AI-assisted surveillance planning

OpenAI banned likely China-origin accounts using AI to draft surveillance-tool pitches, analyze documents, and debug code.

Deceptive Employment Scheme: AI-assisted hiring deception

OpenAI banned accounts potentially used to facilitate a deceptive employment scheme with characteristics of publicly reported North Korea-linked IT-worker activity.

Operation “Wrong Number”: AI-assisted task scam

OpenAI banned accounts that likely originated in Cambodia and used AI to support scam workflows targeting people in the UK.

Univé builds an AI-ready workforce

See how Univé built an AI-ready workforce with ChatGPT Enterprise by combining leadership, responsible governance, and employee-led innovation to transform work at scale.

Advancing responsible AI across Europe

OpenAI shares how its safety, security, transparency, and provenance practices support responsible AI governance in Europe. The work will continue as the EU AI Act advances.

Concrete AI safety problems

We (along with researchers from Berkeley and Stanford) are co-authors on today’s paper led by Google Brain researchers, Concrete Problems in AI Safety. The paper explores many research problems around ensuring that modern machine learning systems operate as intended.

Machine Learning Unconference

The latest information about the Unconference is now available at the Unconference wiki, which will be periodically updated with more information for attendees.

Attacking machine learning with adversarial examples

Adversarial examples are inputs to machine learning models that an attacker has intentionally designed to cause the model to make a mistake; they’re like optical illusions for machines. In this post we’ll show how adversarial examples work across different mediums, and will discuss why securing systems against them can be difficult.

Interpretable machine learning through teaching

We’ve designed a method that encourages AIs to teach each other with examples that also make sense to humans. Our approach automatically selects the most informative examples to teach a concept—for instance, the best images to describe the concept of dogs.