Veille scientifique · Page 5

Democratic governance of frontier AI, part II: Empowering cyber defenders

Evaluation & safety · provider_announcement

OpenAI outlines a blueprint for empowering cyber defenders with frontier AI, stronger safeguards, modernized cybersecurity, and more resilient public institutions.

Patch the Planet: global response coordination as agents uncover critical vulnerabilities

Agent systems · provider_announcement

Patch the Planet connects security researchers, maintainers, and incident responders to help address critical vulnerabilities across open source.

Inside NVIDIA Sentry: Secure AI Agents at Runtime

Agent systems · primary_technical_publication

An AI agent can complete its assignment and still cross a security boundary. Recent incidents show that agents can circumvent security controls while using...

Design Proteins with RFD4-Proteina Using Reward-Guided Inference-Time Search

Compute & efficiency · primary_technical_publication

Protein design is a search problem. Choose amino acids for an eight-residue protein sequence and more than 25 billion combinations are possible. Choose amino...

OpenAI models are coming to Google Cloud

Agent systems · provider_announcement

OpenAI and Google Cloud are bringing frontier models and Codex workflows to Gemini’s Model Garden, available through Google Cloud Marketplace.

Achieve Productive AI Capacity Faster and More Predictably with NVIDIA DSX POD

AI models · primary_technical_publication

AI factories turn energy into intelligence. But before an AI workload can produce its first token, operators must bring an interconnected system of compute,...

How NVIDIA OpenShell Reference Architecture Secures and Governs AI Agent Workloads at Scale

Agent systems · primary_technical_publication

Autonomous agents can work for hours on complex tasks, accessing files, tools, the internet, and spawning other agents to collaborate with. Executing complex...

Solving a 31-year-old grand challenge in mathematics – the KLS conjecture

Mathematical reasoning · primary_technical_publication

For decades, a fundamental question about the difficulty of computing in high-dimensional spaces has resisted proof. New research resolves this question, removing a major theoretical

Using thousands of AI agents to uncover new leads in biological data

Agent systems · provider_announcement

See how OpenAI, PerturbAI, and Scripps Research used distributed AI agents to uncover promising research leads in data from millions of brain cells.

How Specializing the Vocabulary of SLMs Can Cut Inference Cost by 77% in Real Applications

Compute & efficiency · primary_technical_publication

Developing an AI Factory Digital Twin with an Agent-Assisted Workflow

Agent systems · primary_technical_publication

Building an AI factory digital twin requires developers to bring site plans, equipment assets, and engineering requirements into a consistent model that teams...

Terminal X matches senior investor judgment 88.6% with GPT-6.1

Agent systems · provider_announcement

Terminal X’s Eval Agent grades outputs against a 54-criterion rubric, catching 25% more deviances from a fund’s mandate with GPT-6.1 Sol.

Maestro: Agentic Formal Proof Development by Coordination

Formal verification · preprint

Lean is increasingly used to verify mathematical proofs and software produced by LLM agents, but writing formal proofs remains hard. Existing prover systems take a

Advancing coding in low‑resource programming languages

Evaluation & safety · provider_announcement

Learn how OpenAI is improving ChatGPT across low-resource programming languages, with new evaluations for software engineering, SQLite, and COBOL migration.

MAMath-Bench: Process-Aware and Robust Evaluation of Multimodal Multi-Agent Mathematical Reasoning

Mathematical reasoning · preprint

Mathematical reasoning is a fundamental capability for educational applications of Large Language Models (LLMs), where reliable problem solving requires not only correct answers but also

How Localiza reorganized work around generative AI

Agent systems · provider_announcement

See how Localiza uses ChatGPT Enterprise, the OpenAI API, Codex, and 60+ specialized agents to reorganize work and build AI-native mobility experiences.

Designing more efficient quantum circuits with Claude

Complexity theory · provider_announcement

Working largely on its own, Claude designed quantum circuits that improve on the best published versions, in some cases by an order of magnitude or

Stack Overflow for Agents triples signups with ChatGPT plugin

Agent systems · provider_announcement

Two engineers shipped Stack Overflow for Agents in a few weeks with Codex, and its ChatGPT plugin helped drive up to 10× growth in agent

How Itaú is building generative AI for its entire customer base

AI models · provider_announcement

With ia.i, a conversational interface, and the IARA governance platform, Itaú is applying its AI First strategy to customers and employees — in one of

How Grupo Boticário uses AI to strengthen human connections

Agent systems · provider_announcement

Almost 50 years after starting in a compounding pharmacy in Paraná, the group operates a proprietary agent platform that autonomously resolves up to 80% of

Code, Test, and Deploy on One Machine with NVIDIA RTX Spark

Agent systems · primary_technical_publication

AI agents are changing how software gets built. Developers have gone from writing every line of code by hand to directing agents that write most

Welcome Muse Spark Open: open-weight frontier multimodal mixture-of-experts model from Meta

Multimodal AI · primary_technical_publication

NVIDIA Nemotron 3.5 Super VL Powers Specialized Agents with Multimodal Reasoning

Agent systems · primary_technical_publication

Always-on agents are taking on longer, more complex enterprise workflows. A single request can involve reading a document, checking information across sources,...

Teaching students to think and learn through questions

AI models · provider_announcement

Celebrate Teachers’ Day with AI for Educators in Brazilian Portuguese and see how Study Mode can help teachers support active, step-by-step learning.