Latest AI (Artificial Intelligence) News
KPMG rolls out Claude to 276,000 employees worldwide
KPMG and Anthropic announced a broad deployment of Claude through KPMG Digital Gateway, reaching 276,000 employees across 138 countries. The rollout begins with tax and private equity use cases and is set to expand across advisory services, with full implementation on Microsoft Azure planned by September 2026.
OpenAI launches DeployCo, a $4 billion AI consulting subsidiary
OpenAI has formed a majority-owned consulting company called DeployCo, backed by more than $4 billion from a consortium of 19 investors and partners. The move highlights how AI companies are increasingly competing not just on models, but on controlling enterprise deployment and implementation.
Cohere acquires Aleph Alpha to build a transatlantic sovereign AI challenger
Cohere’s acquisition of Aleph Alpha is being framed as a bid to create a $20 billion sovereign AI competitor spanning North America and Europe. The deal signals continued consolidation in the enterprise and government-focused AI market.
Canada rules ChatGPT violated privacy law
Canada’s privacy authorities ruled that OpenAI violated national privacy laws, adding another major regulatory challenge for the company. The case underscores growing global scrutiny of how AI systems collect, store, and use personal data.
Microsoft Build expected to showcase multi-model Azure AI Foundry
Ahead of Microsoft Build, reports point to an Azure AI Foundry update that would bring Anthropic Claude alongside OpenAI models in the same platform. The event is also expected to feature multi-agent orchestration in GitHub Copilot and new audit capabilities for enterprise use.
Bearly AI ships OpenADE with GPT-5.5, Codex, and Claude Code support
Bearly AI released OpenADE, an AI coding workflow tool designed to support multiple frontier coding assistants, including GPT-5.5, Codex, and Claude Code. The launch reflects continued competition in developer tools centered on orchestration rather than a single model.
OpenAI updates GPT-5.5 Instant with stronger factuality and personalization
OpenAI released GPT-5.5 Instant as the default ChatGPT model, emphasizing improved factual accuracy, fewer hallucinations, and stronger personalization from user context. The update suggests a shift toward more reliable everyday assistant behavior rather than only benchmark gains.
Google announces Gemma 4 speed improvements with Multi-Token Prediction
Google said its Gemma 4 open-model family has surpassed 60 million downloads and introduced Multi-Token Prediction drafters to speed up inference. The company says the change can deliver up to a 3x speedup without degrading reasoning quality.
Subquadratic debuts a 12-million-token context window
Subquadratic unveiled a model with a 12-million-token context window, one of the largest publicly discussed context capacities so far. The release pushes the boundaries of long-context retrieval, code analysis, and document processing.
Skild AI explains its 'OmniBrain' universal robotics foundation model
Skild AI described Skild Brain as an “OmniBrain” designed to generalize intelligence across robot types and tasks. The company’s framing indicates a push toward a single foundation model for many physical embodiments rather than task-specific robotics systems.