Daily AI News

OpenAI raises over $4 billion for new enterprise deployment venture

OpenAI has secured more than $4 billion in funding to launch "The Deployment Company," a joint venture aimed at accelerating enterprise AI roll‑outs and services.

Read source →

Anthropic and OpenAI launch joint ventures for enterprise AI services

Both Anthropic and OpenAI announced separate partnerships with asset managers to create dedicated enterprise AI service outfits, targeting large‑scale corporate customers.

Read source →

DoorDash adds AI tools to speed up merchant onboarding and edit dish photos

DoorDash rolled out a suite of generative‑AI features that automate merchant onboarding, enhance food‑photo aesthetics, and auto‑generate storefront websites.

Read source →

Google launches Gemini Enterprise Agent – native agentic AI governance product

At Google Cloud Next ’26, Google introduced Gemini Enterprise Agent, a built‑in governance layer for agentic AI, enabling enterprises to enforce policies, audit usage, and manage risk.

Read source →

OpenAI unveils Symphony – a self‑orchestrating agent system to reduce human attention bottleneck

OpenAI announced Symphony, a framework where multiple AI agents autonomously pull tickets from Linear, execute code, and report results without constant human supervision.

Read source →

Zyphra introduces Tensor and Sequence Parallelism (TSP) for 2.6× throughput gains

Zyphra’s TSP technique combines tensor‑level and sequence‑level parallelism, cutting memory usage and delivering up to 2.6× speed‑up for large‑scale model training and inference on the same GPU axis.

Read source →

Open models have crossed a performance threshold, matching closed‑frontier models on core agent tasks

Evaluation of open‑weight models such as GLM‑5 and MiniMax M2.7 shows they now rival proprietary models in file operations, tool use, and instruction following while costing far less.

Read source →

Open SWE: an open‑source framework for internal coding agents

Open SWE provides reusable components for building internal code‑generation agents, built on Deep Agents and LangGraph, and released under an Apache‑2.0 license.

Read source →

Image‑AI models now drive app growth, beating chatbot upgrades

Appfigures data shows visual‑model releases generate 6.5× more app downloads than chatbot updates, though conversion to revenue remains modest.

Read source →

7 practical ways to reduce Claude Code token usage

The article outlines tactics—prompt pruning, context compression, and selective tool invocation—to lower token consumption when using Claude Code for programming assistance.

Read source →