01 Updates
New Gemini tools for small business owners Look for these new tools to roll out globally in the coming days to help you save time and focus on your business.
Check out "New Gemini tools for small business owners" to simplify your daily work.
Source-linked proof · original source 02 Updates
Work management context for AI agents with Smartsheet MCP on AWS Smartsheet has launched a generally available Model Context Protocol (MCP) Server.
The AWS Machine Learning Blog release records generally available for Amazon.
Source-linked proof · original source 03 Updates
GitHub package firewall in public preview There's been a recent flood of malware attacks that exploit systems at package installation time.
The GitHub Changelog release records public preview for GitHub.
Source-linked proof · original source 04 Updates
OpenAI: You can now choose 12-hour prompt caching in the API. You can now choose 12-hour prompt caching in the API.
The X @OpenAIDevs page documents the OpenAI update.
Source-linked proof · original source 05 Updates
Agent Skills Agent Skills are on-demand instructions, resources, and scripts.
The Cloudflare Developer Docs page documents the Cloudflare update.
Watch next Watch the Cloudflare release for patch notes and rollout issues. Source-linked proof · original source 06 Updates
Powering the Agent Development Lifecycle with any framework, from ADK to LangChain and beyond For teams building with Agent Development Kit (ADK), that opinionated path has worked.
The Google Developers Blog page documents the Google update.
Watch next Watch the Google release for patch notes and rollout issues. Source-linked proof · original source 07 Updates
Access OpenAI GPT models on Amazon Bedrock from Canada with cross-Region inference With Amazon Bedrock cross-Region inference (CRIS), you can now access these models from Canada.
The source records 6.1 / 5.6.
Source-linked proof · original source 08 Updates
Build a semantic layer for agentic AI on AWS with AtScale and Amazon Bedrock AgentCore AtScale takes a different shape from the first two.
The semantic model is dimensional (dimensions, hierarchies, and measures) rather than a graph of triples or a set of views.
09 Updates
Build a semantic layer for agentic AI on AWS with Denodo and Amazon Bedrock AgentCore Amazon Bedrock AgentCore is a platform to build, connect, and optimize agents at scale, with any framework or model.
The AWS Machine Learning Blog page documents the Amazon update.
Source-linked proof · original source 10 Updates
FairStress: A deeper evaluation of fairness in LLM decision-making Standard demographic fairness benchmarks often test whether a model selects the same candidate in otherwise identical scenarios after the candidates' demographic identities are swapped.
The Hugging Face Blog page documents the Ai2 update.
Source-linked proof · original source 11 Updates
From data to decisions: Conversational supply chain intelligence with Amazon Quick and Amazon Bedrock AgentCore Could that manager instead ask in plain language and get an actionable, cross-system answer in seconds?
Amazon Bedrock AgentCore is the intelligence layer, exposing enterprise Model Context Protocol (MCP) servers as a composable tool layer.
12 Updates
Announcements Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
The Anthropic page documents the Anthropic update.
13 Updates
ALTK-Evolve: How learned memory tops the 2-bench leaderboard on customer service tasks without a frontier model Those guidelines are then added to the agent's prompt for subsequent conversations.
The source records 4.1 / GPT-4.
Source-linked proof · original source 14 Updates
Get hands-free personalized help on the road with Gemini To get started, check your Gemini settings to opt in and connect your preferred apps.
The blog.google page documents the Google update.
Source-linked proof · original source 15 Updates
Anthropic: New on our Science Blog: designing protein binders against scorpion neurotoxins. New on its Science Blog: designing protein binders against scorpion neurotoxins.
The X @AnthropicAI release records 1.2m for Anthropic.
Source-linked proof · original source 16 Updates
OpenAI: Keep sensitive conversations out of view with Locked Chats in ChatGPT. Keep sensitive conversations out of view with Locked Chats in ChatGPT.
The X @ChatGPT page documents the OpenAI update.
Source-linked proof · original source 17 Updates
Advancing digital sovereignty with greater control and choice Organizations are considering more than where data resides.
The Microsoft Official Blog page documents the Microsoft update.
Source-linked proof · original source 18 Updates
Build a semantic layer for agentic AI on AWS with Strategy Mosaic and Amazon Bedrock AgentCore The second post kept the same customer 360 (C360) pattern and swapped in Denodo's data virtualization platform, extending the source mix to include an on-premises PostgreSQL cluster.
The AWS Machine Learning Blog page documents the Amazon update.
Source-linked proof · original source 19 Updates
Securing AI agent identity with X.509 PKI on AWS Traditional approaches such as sharing API keys between components or passing OAuth tokens through agent chains fall short of the zero-trust requirements that enterprise deployments demand.
The AWS Machine Learning Blog page documents the Amazon update.
Source-linked proof · original source 20 Updates
Deep dynamical models of single-cell multiomic velocities predict loss-of-function and rescue perturbations in B cells Perturbing Arid1aHet GCs in silico identified Ctcf, Stat3 and Bcl6 as rescue targets for Arid1a loss.
The Nature Machine Learning page documents the Nature update.
Source-linked proof · original source 21 Updates
From conversation to payment: Sierra and Stripe Payments are becoming increasingly important, as agents move from one-and-done support conversations to interactions that can drive revenue.
The Sierra Blog page documents the Sierra update.
Source-linked proof · original source 22 Updates
Broad access through consistent standards Models such as GPT-6 Astra expand what developers and businesses can build.
The OpenAI page documents the Openai update.
Source-linked proof · original source 23 Updates
A curriculum for post-training agents These stages turn a pretrained model into one that follows instructions, uses tools or solves a specific task.
The Hugging Face page documents the Huggingface update.
Watch next Watch the Huggingface release for patch notes and rollout issues. Source-linked proof · original source 24 Updates
Fine-tuning GPT-OSS-20B for multilingual reasoning on Amazon SageMaker HyperPod It raises a question every team faces once they start customizing open-weight models: supervised fine-tuning (SFT), reinforcement learning with Group Relative Policy Optimization (GRPO), or both?
The AWS Machine Learning Blog page documents the Amazon update.
Source-linked proof · original source 25 Updates
Beyond the KV cache: Training and serving i14's Laplace 3B linear attention model on AWS Trainium2 with native PyTorch This preserved its Gated Linear Attention (GLA) design and Muon optimizer.
The AWS Machine Learning Blog release records 29.5 for Amazon.
Source-linked proof · original source 26 Updates
OpenAI: You can now automate recurring engineering work with team tasks in ChatGPT, like pulling together what shipped and whats blocked ev... You can now automate recurring engineering work with team tasks in ChatGPT, like pulling together what shipped and what's blocked every Friday.
The X @OpenAIDevs page documents the OpenAI update.
Source-linked proof · original source 27 Updates
Ask a Scientist: How can AI help doctors spend more time on collaborative care? The team is now expanding the project to include audio and video interactions while preparing for larger-scale clinical studies.
The blog.google page documents the Google update.
Source-linked proof · original source 28 Updates
Google Gen AI SDK releases 2.32.0 The pypi registry records Google Gen AI SDK version 2.32.0.