The Archive

Deepening our partnership with the UK government to support prosperity and security in the AI era

OpenAI details cybersecurity risk mitigation and defensive capabilities as AI model power increases.

Anthropic donates Model Context Protocol and establishes Agentic AI Foundation for agent standards.

Scout24 built GPT-5 conversational real-estate assistant with clarifying questions and personalized listing recommendations.

Accenture and Anthropic announce multi-year partnership to productionize enterprise AI deployments.

Mistral ships Devstral 2 and Mistral Vibe CLI, open-source agentic coding models with autonomous agent scaffolding.

Systematically evaluating the factuality of large language models with the FACTS Benchmark Suite.

OpenAI co-founds Agentic AI Foundation under Linux Foundation; donates AGENTS.md standard for interoperable agentic AI.

OpenAI launches certification courses and AI Foundations training to build workforce skills.

OpenAI and Deutsche Telekom deploy ChatGPT Enterprise across European telecom infrastructure.

Commonwealth Bank of Australia rolls out ChatGPT Enterprise to 50,000 employees for customer service and fraud detection.

Denise Dresser appointed OpenAI Chief Revenue Officer overseeing enterprise and customer strategy.

Import AI 437 examines co-improvement mechanisms, RL-based dreaming, and human labeling efficiency in AI systems.

OpenAI integrates grocery shopping and checkout functionality into ChatGPT via Instacart partnership.

OpenAI reports accelerating enterprise AI adoption, deeper integration, and measurable productivity gains in 2025.

Virgin Atlantic uses AI to accelerate development, improve decision-making, and enhance customer experience.

OpenAI establishes sovereign AI infrastructure in Australia and commits to upskilling 1.5M workers.

Scientists are using AlphaFold to strengthen a photosynthesis enzyme for resilient, heat-tolerant crops.

Snowflake and Anthropic partner on $200M agentic AI initiative targeting enterprise customers.

OpenAI acquires Neptune to enhance model interpretability, experiment tracking, and training monitoring capabilities.

OpenAI researchers test 'confessions' training method to improve language model honesty and error acknowledgment.

OpenAI Foundation distributes $40.5M in unrestricted grants to 208 nonprofits for community AI innovation.

Anthropic acquires Bun JavaScript runtime; Claude Code reaches $1B valuation milestone.

Mistral releases Mistral 3, a family of open-source multimodal frontier models supporting text, image, and audio.

Anthropic launches discounted Claude access for nonprofits with free training and support.

Mirakl integrates ChatGPT Enterprise for agentic commerce, automating documentation and customer support workflows.

OpenAI awards up to $2M in grants for research on AI applications and risks in mental health.