OpenAI has acquired the startup Orchestra Bio, adding its team and technology to its biotech efforts. The acquired capabilities will support Rosalind, OpenAI's AI model for life-sciences research, and Rosalind Workbench, a tool for planning experiments and analyzing results. The deal signals continued investment in applying AI to scientific research workflows.
Meta has dropped the word "train" from its updated commercial terms of service, which take effect October 30, following criticism from legal experts. Experts had warned the sweeping terms could let Meta scrape websites linked on business pages and avoid licensing fees from news organizations by treating business page data as AI training data. A Meta spokesperson confirmed the company did not notify users of the revised language. Meta acknowledged the initial Nieman Lab report but did not address
ElevenLabs has released two speech-to-speech models under the Eleven Realtime name, built for real-time AI conversations. The v1 Max variant reportedly scores 98.8 percent on the Big Bench Audio benchmark with a 0.57-second response time. ElevenLabs claims no model on the Artificial Analysis leaderboard is both faster and more accurate.
Deepseek has open-sourced DeepInfer, the inference framework powering its fast V4.1-Flash model. The vLLM-based system is described as reducing compute costs and energy use when running AI models. The release is framed as adding pricing pressure on Western providers such as OpenAI and Anthropic.
Meta has paused a Muse feature that let users share conversations with its AI assistant after security researchers at Claroc found more than 6,800 Muse links exposed through public DNS records. Meta spokesperson Tracy Clayton confirmed the sharing functionality creates link-accessible pages and said it is paused pending a review of the report and possible changes to link behavior. No timeline for restoring the feature was given.
AWS published an end-to-end walkthrough of a conversational supply-chain intelligence pattern in which Amazon Quick serves as the user interface and Amazon Bedrock AgentCore provides the agent runtime, gateways, identity and policy layer. Business systems such as disruption alerts, governance, quoting and invoicing are exposed as MCP servers through AWS Agent Registry, letting an agent plan multi-step tasks and return consolidated impact assessments and mitigation options. The post includes an o
AWS details a fine-tuning workflow on Amazon SageMaker HyperPod that teaches GPT-OSS-20B to reason and answer in French, German, Spanish and Italian instead of defaulting to English. Supervised fine-tuning with LoRA raised reasoning-language accuracy from 22% to 100% on a 50-case evaluation, while adding GRPO on top of the SFT checkpoint lifted answer-language accuracy from 62% to 70%. A controlled Qwen2.5-7B experiment across multilingual reasoning and GSM8K tasks supports the argument that GRP
Why it matters: It offers a concrete, reproducible fine-tuning methodology and decision framework for choosing between SFT and GRPO when adapting open-weight reasoning models to non-English languages.
A magistrate judge has ordered Meta to identify executives who approved downloading books from shadow libraries like Anna's Archive and LibGen for AI training data, according to TorrentFreak. The discovery order affects a class action by authors including Richard Kadrey and Sarah Silverman and a separate suit from publisher Entrepreneur Media. Judge Thomas Hixon largely sided with plaintiffs seeking the decision-making chain of command, after Meta argued its dataset collection was managed throug
Octave's Reliance quality management platform built a shared three-layer AI foundation on Amazon Bedrock to serve regulated-industry customers while keeping tenant data isolated. The architecture uses Cohere Embed v4 for multimodal, multilingual embeddings written to per-tenant Amazon OpenSearch Service indexes, Bedrock AgentCore for isolated per-session execution and scoped credentials, and Bedrock Guardrails applied at input and output. An event-driven pipeline embedding roughly 2 billion docu
Mistral AI released Voxtral 3, a family of language models designed for real-time voice agents. The stack combines speech-to-text, text-to-speech and a voice agent SDK, supports more than 50 languages, and allows agents to use external tools during live conversations. It can be deployed on self-hosted infrastructure or via EU-hosted APIs for greater data control.
AWS announced that Canadian organizations can access OpenAI GPT-5.6, GPT-6 and GPT-6.1 models on Amazon Bedrock from the Canada (Central) and Canada West (Calgary) Regions using cross-Region inference profiles. Requests are invoked from the source Region and routed to destination Regions in North America (US geographic profiles) or supported commercial Regions (global profiles). The post details IAM and SCP permissions, invocation through the OpenAI Responses, Chat Completions and Bedrock Conver
OpenAI improved its latest models for low-resource programming languages including COBOL, Fortran and Visual Basic. On the new MultiLangBench, GPT-6 Astra reportedly reaches an 87.4 percent test pass rate across 20 languages, up from about 38 percent for GPT-5.6 Terra. For COBOL migration, GPT-6.1 Sol is said to reach up to 97.9 percent accuracy, which could matter for modernizing critical financial infrastructure.
AWS published a technical walkthrough of AWS Agent Registry, a managed service for cataloging and discovering MCP servers, agents, and agent skills at runtime. An orchestrator on Amazon Bedrock AgentCore runtime searches approved registry records via hybrid keyword and semantic search, constructs protocol-specific clients, and executes with request-scoped tools. The sample demonstrates MCP-only, MCP plus A2A, and pure semantic-match scenarios, with authentication kept separate from discovery.
An AWS Machine Learning Blog post describes a reference architecture for giving AI agents verifiable, revocable cryptographic identities using Keyfactor EJBCA X.509 PKI on AWS. The design adds signed request envelopes with nonces and timestamps, user-signed delegation assertions, RFC 8693 audience-scoped tokens, and real-time OCSP revocation checks across three trust boundaries: user-to-system, agent-to-agent, and agent-to-MCP-tool-server. It uses Amazon Bedrock with the Strands Agents SDK, Keyc
Eon replaced per-tool credentials and local Model Context Protocol servers with a single authenticated MCP gateway built on Amazon Bedrock AgentCore, giving developers and an autonomous SRE agent unified access to investigation tools across AWS, GCP and Azure. Cedar policies at the gateway enforce least-privilege separation between human and agent identities, and IAM cross-account role assumption removed the need for VPC peering per account. Eon says onboarding dropped from roughly two days to a
AWS's ML blog describes a reference architecture that pairs the Denodo Platform with Amazon Bedrock AgentCore to give agents governed access to federated enterprise data. The Denodo MCP Server exposes semantic views to a Strands agent through AgentCore Gateway, while AgentCore Identity performs on-behalf-of token exchange so end-user identity and Denodo row and column security policies are enforced at query time. A customer 360 example federates Amazon Aurora, Amazon Redshift and an on-premises
AWS published a walkthrough for giving agentic AI agents a governed semantic layer, using Strategy Mosaic over Amazon Redshift and Amazon Aurora with a Strands agent hosted on Amazon Bedrock AgentCore. Mosaic exposes one dimensional model through MCP to agents and through JDBC/ODBC, XMLA, and REST to other clients, with linked models handling cross-source joins without ETL. The post covers metric definitions, named filter objects, hierarchies, and propagation of per-user row-level and object-lev
AWS's latest semantic-layer post in its series swaps AtScale in for Stardog and Denodo, modeling a Customer 360 domain in AtScale's open YAML-based Semantic Modeling Language versioned in Git. A single deployed AtScale model is served simultaneously over SQL, XMLA and MCP, letting a Strands agent hosted on Amazon Bedrock AgentCore and existing BI tools get the same computed answers from shared metric definitions. The article details query planning and pushdown to Amazon Redshift Serverless, auto
