News
Updates, product notes and practical coverage for AI platforms and operators.
Google says its News AI pilot now includes more than 400 publications worldwide. The licensing program supports AI article overviews on publisher pages in Google News and customisable audio briefings, with attribution and links back to reporting. The expansion brings major and local publishers into a larger test of AI-assisted news discovery and commercial participation.
Google has updated its generative-AI guardrails guidance to explain a user-controlled Workspace connection. When enabled, selected Workspace data can personalise other Google services and, depending on settings, may be used to train generative AI models with limited human review. The clarification makes consent, account controls and organisational policy central to data governance.
Google Cloud Consulting has introduced SAVE, a four-part workflow for more reliable generative media. It separates factual research from storytelling, anchors generation in brand references, validates outputs through structured review and treats creative production as an evaluated system. The approach shifts attention from elaborate prompts to evidence, constraints and repeatable quality control.
Grok 4.6 is now generally available through Amazon Bedrock, according to xAI. Supported AWS-region customers can access the flagship model through Bedrock with a 500,000-token context window, configurable reasoning effort and listed pricing of US$2 per million input tokens and US$6 per million output tokens.
xAI has made Imagine Image 2.0 generally available as Grok's new Quality Mode, adding targeted edits, segmentation, background removal, multi-reference inputs and smart resizing. The release reaches Grok's web and mobile products now, with reusable creative templates included, while API access remains a future addition.
Scenario has released coordinated API, compute and web-app updates that tighten project-level model control while expanding video, 3D and workflow options. The 6 August package adds Grok and MiniMax video models, Meshy smart-topology assets, custom Cartwheel characters, richer Gaussian splats and more reliable moderation and automation controls.
You.com has added an Answer API that returns a web-grounded Markdown response with inline citations through one endpoint. It performs a single search, checks cited claims against source text and exposes the pages considered. At US$5 per 1,000 calls, it targets applications that need quick answers without building a retrieval pipeline.
Black Forest Labs has released FLUX.3 Video through its API and selected partners, combining video generation, native audio and editing controls in one model family. The launch supports clips of up to 20 seconds, multiple shots, keyframes and continuation from an existing video segment.
Midjourney has released its V8.2 image model, targeting stronger aesthetics, fewer low-quality outputs and better personalisation. The company has not disclosed benchmark data, pricing or rollout details, so users should treat the release as an invitation to test creative consistency rather than a fully documented performance upgrade.
Meshy has added native 8K base-colour texture generation for 3D assets using a new Multiview Diffusion pipeline. The paid-plan feature targets close-up game, film and large-format work, while creators should still test material consistency, processing cost and downstream engine performance.
OpenAI has launched GPT-5.6 across ChatGPT, Codex and its API. The new family includes the flagship Sol model, the balanced Terra model and the lower-cost Luna model, alongside new multi-agent capabilities for completing demanding professional work.