ChatGPT Pulse : l’assistant qui travaille pendant votre sommeil
OpenAI launches Pulse, a preview feature that makes ChatGPT proactive. The assistant can work autonomously in the background, completing tasks without user intervention.
69 articles
OpenAI launches Pulse, a preview feature that makes ChatGPT proactive. The assistant can work autonomously in the background, completing tasks without user intervention.
Spendesk and CFO Connect release a 2025 CFO tools benchmark showing growing AI adoption in finance departments. The study maps CFO practices and preferences regarding automation and artificial intelligence solutions.
OpenAI releases Sora 2, a video-audio generation model with improved physics accuracy, sharper realism, synchronized audio, enhanced steerability, and expanded stylistic range. Direct successor to Sora addressing longstanding challenges in video synthesis.
OpenAI releases Sora 2 and the Sora app with safety built-in to address risks from an advanced video model and social creation platform. The approach relies on concrete protections (specific technical details not disclosed in excerpt).
OpenAI releases Sora 2, its latest video generation model with improved physical accuracy, realism, and control. Features synchronized dialogue and sound effects. Available in the new Sora app.
OpenAI uses AI to enhance customer support operations: faster response times, improved quality, and scaling capacity for rapid growth. No specific metrics or models detailed in excerpt.
OpenAI launches "OpenAI on OpenAI" series documenting how the company uses its own models to streamline operations, scale expertise, and sharing lessons for other organizations to adopt similar practices.
OpenAI built a system to extract contract data quickly, reducing turnaround times and improving team access to contract details. No specific technical details or model names are disclosed in the excerpt.
OpenAI deploys an internal research assistant to analyze millions of support tickets, accelerate insight extraction, and scale investigation capacity across the organization.
OpenAI rolling out parental controls and parent resource page to help families manage ChatGPT usage at home. No technical details or timeline specified in announcement.
OpenAI strengthens its fight against online child sexual exploitation through strict usage policies, advanced detection tools, and cross-industry collaboration. The company blocks illegal content, reports abuse to authorities, and prevents AI misuse.
OpenAI launches agentic commerce protocol in ChatGPT, enabling users, AI agents, and businesses to transact together with instant checkout. First step toward integrating transactional commerce into AI agents.
Hugging Face introduces VibeGame, a vibe-based coding game platform. The initiative explores how to gamify code learning through playful mechanics and progressive challenges.
Hugging Face accelerates Qwen3-8B agent inference on Intel Core Ultra using depth-pruned draft models. The technique reduces inference latency while maintaining response quality for agentic tasks.
Alibaba unveils Qwen-3 Omni, the next iteration of its multimodal model. Following Qwen-2.5 Omni released in March, this version continues Alibaba's acceleration in omnimodal capabilities integrating text, audio, and vision.
The European Trustworthy AI Association (ETAIA) establishes itself as a key player in structuring Europe's AI ecosystem at the Trustworthy AI Summit. The association promotes industrial and responsible AI aligned with European regulations.
Nvidia and Hugging Face release Nemotron-Personas-Japan, a synthetic dataset for training sovereign Japanese AI models. The dataset includes generated personas and dialogues to enhance local model capabilities without relying on foreign infrastructure.
OpenAI and AARP partner to help older adults stay safe online through AI training, scam-detection tools, and nationwide programs via OpenAI Academy and Senior Planet. Social responsibility initiative with no specific technical details disclosed.
Swift Transformers reaches version 1.0. The library for language models on Apple Silicon provides native optimizations and CoreML integration. Future roadmap includes multimodal support and performance improvements.
OpenAI introduces shared projects, improved connectors, and compliance/security updates to ChatGPT for team collaboration. No technical details or metrics disclosed in the announcement.
OpenAI introduces GDPval, a new evaluation framework measuring model performance on economically valuable real-world tasks across 44 occupations. The benchmark assesses practical capabilities on professional use cases rather than traditional academic benchmarks.
OpenAI releases ChatGPT Pulse preview for Pro users on mobile. The feature enables ChatGPT to proactively research and deliver personalized updates based on user chats, feedback, and connected apps like calendar integration.
ENEOS Materials deploys ChatGPT Enterprise to accelerate R&D, improve plant safety design, and cut HR analysis time by 90%. 80% of users report improved workflows.
Siemens and Altair announce a strategic alliance to develop industrial AI in Europe. Jean-Marie Saint-Paul (Siemens) and François Weiler (Altair) discuss the stakes of this partnership aimed at strengthening European technological competitiveness.
Arcep releases a 2025 survey on the environmental impact of digital technology in France. Sustainability becomes a strategic issue for regulators, companies, and society.
MongoDB enables vector search on self-managed editions. Announced September 19 at MongoDB.local NYC, the feature extends vector search capabilities beyond cloud offerings to self-hosted deployments.
The Charles Sadron Institute (CNRS) and Alysophil establish ACTIVIAFLOW, a joint laboratory to develop sustainable chemical processes using AI. This research-industry partnership aims to optimize chemical reactions and reduce waste.
SAP and OpenAI launch "OpenAI for Germany," a 2026 partnership to deliver sovereign, secure AI to Germany's public sector. Regulatory compliance and digital sovereignty initiative for government services.
OpenAI, Oracle, and SoftBank announce five new Stargate AI datacenter sites to accelerate a $500B, 10-gigawatt U.S. infrastructure buildout, creating tens of thousands of jobs.
Hugging Face introduces Smol2Operator, a post-trained GUI agent for automating computer tasks. The model uses vision-based approach to interpret user interfaces and execute actions autonomously.
CNA transforms its newsroom with AI, according to Editor-in-Chief Walter Fernandez in an OpenAI blog post. The article explores AI adoption, organizational culture, and journalism's future without specifying tools or measurable outcomes.
SchoolAI launches an AI platform using GPT-4.1, image generation, and TTS serving 1+ million classrooms. Teacher-controlled tools designed for oversight, student engagement, and personalized learning.
OpenAI and NVIDIA announce strategic partnership to deploy 10 gigawatts of AI datacenters powered by NVIDIA systems, first phase launching in 2026.
Hugging Face introduces SyGra, a unified framework for generating training data for LLMs and SLMs. The tool automates synthetic dataset creation and enables quality optimization across different model sizes.
Hugging Face releases Gaia2 and ARE to enable the community to study AI agents. These tools facilitate research and evaluation of multi-agent systems by providing benchmarks and open-source resources.
OpenAI releases a coordinated vulnerability disclosure policy governing how security flaws in its systems are reported and fixed. The policy establishes a responsible notification process before public disclosure, protecting users and allowing time for remediation.
Figure AI raises over $1 billion in Series C, boosting valuation from $2.6B to $39B. The startup accelerates deployment of its humanoid robots.
Scaleway joins Hugging Face Inference Providers. The integration enables users to access models through Scaleway's cloud infrastructure, expanding deployment options available on the platform.
Hugging Face releases RiskRubric.ai, an open-source platform for assessing AI model safety risks. The tool provides standardized, reproducible metrics enabling developers to identify and mitigate vulnerabilities before deployment.
Apollo Research and OpenAI developed evaluations to detect hidden misalignment ("scheming") in AI models. Behaviors consistent with scheming were observed in controlled tests across frontier models. The team proposes an early method to reduce this phenomenon.
Hugging Face introduces a new 'Public AI' category on its inference platform, enabling developers to access open-source models through a unified API. The initiative aims to democratize AI model access without reliance on proprietary providers.
Replicate introduces a new search API to find the best models and collections with a single API call.
OpenAI announces Stargate UK, an AI infrastructure initiative in the United Kingdom. No technical details or timeline provided in available excerpt.
OpenAI is integrating age prediction and parental controls into ChatGPT to tailor content for minors. The feature aims to provide age-appropriate, safer experiences with new family-oriented tools.
Hugging Face releases LeRobotDataset v3.0, a major update bringing large-scale datasets to the lerobot platform for training robotics models. The version includes new data, performance optimizations, and improved resource accessibility.
OpenAI upgrades Codex with faster performance, improved reliability, and real-time collaboration. The tool now works across terminal, IDE, web, and mobile for autonomous task execution.
OpenAI releases research from its largest ChatGPT usage study, showing the tool creates economic value across personal and professional contexts. Adoption is expanding beyond early users, integrating AI into everyday life. No specific metrics or findings are detailed in the excerpt.
OpenAI releases GPT-5-Codex, a GPT-5 variant optimized for agentic coding. The model dynamically adjusts thinking effort based on task complexity: fast responses for simple queries, extended reasoning for complex tasks.
Hugging Face integrates visible watermarking in Gradio to protect generated images. The feature adds a discrete but detectable watermark to outputs, useful for tracing the origin of AI-generated content.
OpenAI announces partnership with US CAISI and UK AISI to strengthen AI safety and security. No concrete technical details or results provided in excerpt.
Hugging Face introduces Palmyra-mini, a family of lightweight models designed for reasoning tasks. These models deliver strong performance despite their reduced size, optimized for resource-constrained deployments.
OpenAI restructures governance: nonprofit arm gains equity stake in PBC subsidiary, accessing $100B+ in resources for safe AI development. Reaffirms nonprofit control over long-term strategy.
OpenAI and Microsoft sign a new memorandum of understanding (MOU) strengthening their partnership on AI safety and innovation. No financial terms or technical details disclosed in the joint statement.
Hugging Face shares optimization techniques from OpenAI applicable to open-source transformer models. The article details fine-tuning, inference, and training methods reproduced from OpenAI practices to improve model performance.
Hugging Face and Together AI enable fine-tuning any LLM from the Hub through direct integration. Users can adapt open-source models without proprietary infrastructure, with pay-as-you-go billing.
Hugging Face introduces Jupyter Agents, an approach to train LLMs to reason with Jupyter notebooks. The system enables models to execute code, analyze results, and iterate to solve complex problems.
Article positioning AI as normal technology rather than revolutionary. Challenges dominant hype discourse and proposes a more nuanced perspective on actual capabilities and limitations of current systems.
SafetyKit, OpenAI's content moderation tool, leverages GPT-5 to enhance risk detection and compliance enforcement. No performance metrics or comparative benchmarks disclosed in excerpt.
Hugging Face introduces mmBERT, a multilingual extension of ModernBERT. The model extends modern BERT architecture to 101 languages with improved efficiency and performance on classification and similarity tasks.
OpenAI launches a $50M People-First AI Fund for U.S. nonprofits focused on education, community innovation, and economic opportunity. Unrestricted grants available; applications close October 8, 2025.
Replicate implements PyTorch torch.compile caching to reduce boot and inference times. Compiled models are cached across invocations, eliminating recompilation on each run.
OpenAI releases research explaining why language models hallucinate. The study proposes improved evaluation methods to enhance AI reliability, honesty, and safety.
OpenAI and Greek Government launch "OpenAI for Greece" to deploy ChatGPT Edu in secondary schools. The initiative aims to boost AI literacy, support local startups, and drive national economic growth through responsible AI learning.
OpenAI launches a Jobs Platform and Certifications program to connect workers with employment, training, and credentials. Initiative aims to democratize access to AI skills and expand economic opportunity.
Google releases EmbeddingGemma, an efficient embedding model based on Gemma. The model delivers competitive performance with reduced memory footprint, optimized for vector search and RAG tasks.
Hugging Face introduces SAIR, an AI platform to accelerate pharma R&D through intelligent structural analysis. The tool leverages deep learning models to predict molecular properties and optimize molecule design, reducing development cycles.
OpenAI acquires Statsig; Vijaye Raji becomes CTO of Applications reporting to Fidji Simo (CEO of Applications). Integration of experimentation and feature management platform into OpenAI's product stack.
OpenAI adds parental controls for teens, expert partnerships, and routes sensitive conversations to reasoning models in ChatGPT to improve safety and helpfulness.
Hugging Face introduces ahead-of-time (AOT) compilation for ZeroGPU Spaces, enabling faster inference by pre-compiling GPU code. This optimization reduces startup latency and improves overall application performance.