AI progress, without the leaderboard theatre.

StateBench: Testing how language models track state changes over long documents

Agentic cheminformatics with Google Antigravity

Learn how Google Antigravity and our Cheminformatics Skills accelerate drug discovery workflows, search compounds, and propose novel rigidified macrocycles.

Palette: teaching a small model to build decks in your design language

CWE-bench v2: More coverage, quality, and transparency

Transformers Model Integration Contracts

Introducing the Grid Protectors Program

OpenAI’s Grid Protectors Program will work with utilities to help future compute campuses support grid reliability and protect consumer electricity costs.

LifeVal

Introducing LifeVal, a benchmark measuring how well AI helps people complete everyday tasks, from finding childcare to managing finances and government paperwork.

More Rules, Longer Lists? Benchmarking Self-Evolving Agents on a Hidden-Rule Curriculum

arXiv:2610.40758v1 Announce Type: new Abstract: Self-evolving agents rewrite a persistent state from feedback, but it is largely untested whether this lets them learn a rule nobody told them. An agent can also pass by writing down each correction as an excepti

Beyond Plausible Prediction: Safety Sufficiency in Web World Models

arXiv:2610.40798v1 Announce Type: new Abstract: Web world models increasingly predict the consequences of candidate actions before agents act, yet a plausible predicted future does not necessarily support the correct safety decision. In a matched human evaluat

Circumventing the Computational Hardness of Computational Arbitrage

arXiv:2610.40831v1 Announce Type: new Abstract: AI model marketplaces host providers with heterogeneous costs and capabilities. These differences create an opportunity for computational arbitrage (Olmedo et al., 2026), in which an intermediary purchases infere

Incongruence-Aware Evaluation of Multimodal Speech Emotion Recognition in Audio Large Language Models

arXiv:2610.40883v1 Announce Type: new Abstract: Voice-interactive large language models process both what a speaker says and how they say it, yet the two channels do not always agree. We ask whether that disagreement degrades speech emotion recognition. Mappin

ChatGPT as a Surveillance Assistant? An Exploratory Case Study of AI-Assisted Surveillance Risks

arXiv:2610.40890v1 Announce Type: new Abstract: Large language models (LLMs) can generate detailed guidance, functional code, and persuasive text at scale. As one of the most widely used consumer language model products as of 2026, ChatGPT gives broad public a

Does Recursive Knowledge Construction Help Agents Verify Consensus Claims in Science?

arXiv:2610.40906v1 Announce Type: new Abstract: Recent scientific agents build knowledge recursively by retrieving evidence, constructing an intermediate state, and judging whether that state contains sufficient information. These agents are evaluated end-to-e

Towards Portable Specification of the Agent Harness

arXiv:2610.40917v1 Announce Type: new Abstract: Large language models act only through the infrastructure that surrounds them. That infrastructure, the harness, determines what a model can accomplish and how it fails, yet it has no settled definition. To addre

Agentic World Modeling for Agent Test-time Scaling without Environment Replica

arXiv:2610.40927v1 Announce Type: new Abstract: Test-time scaling (TTS) improves large language model agents by sampling and selecting among multiple candidate trajectories, but common realizations assume that these trajectories can be executed in parallel. Th

DiRAFT: Privacy-Aware Financial Transcript Synthesis from Aggregate Discourse Priors

arXiv:2610.40938v1 Announce Type: new Abstract: Using customer conversations to train and evaluate financial language models raises confidentiality concerns because non-public financial facts can remain after direct identifiers are masked. Synthetic transcript

The First Few Tokens Matter: Understanding Chain-of-Thought Controllability

arXiv:2610.40939v1 Announce Type: new Abstract: Chain-of-thought (CoT) monitoring is a promising way to oversee increasingly capable agents. It succeeds in part because current models appear limited in their ability to control what they verbalize in their reas

Benchmarking Forecasting Agents on Yesterday's Tomorrows

arXiv:2610.40940v1 Announce Type: new Abstract: Digital agents can now gather and organize information, reason over it, and use tools to complete complex tasks. Forecasting puts all of these abilities to the test: it requires integrating incomplete evidence, i

Opening the Box Changes the Cat: LLMs Detect Chain-of-Thought Monitoring - What now?

arXiv:2610.40946v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning allows large language models (LLMs) to generate intermediate reasoning traces in natural language while solving complex tasks. These traces can support safety monitoring by exposi

nemotron-halloween

Speeding up planning decisions with AI

Google DeepMind and UK government work together to reduce planning assessment times

Making it easier for websites to express AI training preferences

OpenAI will honor a shared robots.txt signal to help website operators communicate their AI training preferences to participating providers.

Community safety report and new protections for teens

OpenAI is strengthening violence prevention in ChatGPT with expanded safeguards, earlier interventions, and a new commitment to public safety reporting.

Fair use benefits everyone

OpenAI shares excerpts from court briefs by creators, academics, accessibility advocates, and legal experts, among others, on how fair use supports AI training and innovation.

An updated Model Spec for more capable and beneficial AI

OpenAI’s updated Model Spec outlines five core values to guide more capable AI toward sound judgment, human agency, and faithfulness to human intent.

Granite Switch 4.1.1 — Update models like you update software

How we detect frontier model misalignment

OpenAI uses layered monitoring to detect frontier model misalignment and unauthorized agent behavior so responders can intervene before incidents escalate.

How Ropes & Gray is developing AI for transactional diligence

Ropes & Gray is developing AI for transactional diligence with OpenAI. Lawyers review grouped documents, summaries, and reports before advising clients.

Credit Saison achieves 954% ROI in a ChatGPT Enterprise pilot

Credit Saison’s ChatGPT Enterprise pilot delivered 954% ROI. See how leaders and employees helped turn the results into companywide adoption.

Announcing Project ProtoStar: Making Publicly Funded Science AI-Ready

Welcome Le Chonk: Mistral Large 4 on Hugging Face

Introducing AnySpeechRecognizer and AnyDecisionModel: Open Models for Speech and Decisions on Apple Platforms

Gemini 4 Carbon: Our next frontier model

OpenAI and ServiceNow expand partnership across enterprise workflows

OpenAI and ServiceNow expand their partnership to bring OpenAI models and agents into ServiceNow and ServiceNow workflows into ChatGPT and Codex.

The future of the firm

How AI might enter the org chart—beneath, beside, and between us.

Legora is building the agentic operating system for legal work

See how Legora uses OpenAI models, ChatGPT, and Codex to accelerate legal workflows, expand contract analysis, and free lawyers to focus on higher-value work.