11 VERIFIED REPORTS ANCHORED IN TIER 1 RECORDS · SUBSIDIARY OF PRODUCT
All brand marks and logos are the property of their respective owners and used solely to identify the subject of verified news reporting under nominative fair use.
What happened.Microsoft announced Azure Maia 300, its third-generation custom AI accelerator designed for high-density inference and multi-step agentic execution. The hardware uses a multi-chiplet architecture on TSMC's N2P node with 288 GB of HBM4 memory delivering 18 TB/s bandwidth per accelerator.
What happened.Anthropic launched the Anthropic Cyber Mission, establishing the Critical Infrastructure Defense Program to protect operational technology environments using Claude frontier models. Alongside the industrial initiative, Anthropic released the OSS Scanner, a free service that runs security audits across open-source repositories.
What happened.Researchers introduced RSI-Anything, a recursive self-improvement framework that uses predictive world models to guide agent action selection and filter training data. The system simulates candidate trajectories and evaluates predicted outcomes before running actions in real environments.
What happened.OpenAI introduced Sign in with ChatGPT, an authentication protocol that allows third-party software developers to authenticate users and bill API token consumption directly to the end user's personal ChatGPT account or subscription pool. The protocol debuted with 16 integration partners, including Devin, Notion, Vercel, and OpenClaw.
What happened.Barclays announced an expanded strategic collaboration with Anthropic to integrate Claude across its global banking operations. The bank aims for Claude Code adoption to reach 50% of its developer workforce by the end of 2026, rising to a majority of its software engineers by 2027. The rollout builds on existing internal deployments, including an assistant used by more than 16,000 employees in its UK retail division and an email classification system in its Global Markets division processing 120,000 client messages daily.
What happened.OpenAI released the Decisions API alongside Luna, a lightweight model specialized for discrete classification and intent routing that returns responses in sub-second latency. The API requires developers to supply a fixed list of allowable output options in the request payload. The model is constrained to return only choices from the supplied list, bypassing free-form JSON generation.
What happened.Google released Guided Vision within Gemini Live on Android, providing real-time audio descriptions from a live camera feed. Developed alongside blind and low-vision community members with testing by Aira, the system processes continuous video input to describe surroundings and read text without manual photo capture.
What happened.OpenAI released DOTs, a personal agent environment running on GPT-6 Astra priced at $100 per user monthly, capable of autonomous code refactoring, dependency tracking, automated testing, and multi-step pull request generation. OpenAI also established a Pro 500 tier at $500 monthly providing 25 times standard ChatGPT Plus capacity and access to UltraAstra, an accelerated inference mode operating at 300 tokens per second priced in the API at $60 per million input tokens and $300 per million output tokens. The existing Pro $200 plan resumed sales with its monthly quota reduced by half.
What happened.Microsoft AI released MAI-Transcribe-2-Streaming, a low-latency speech-to-text model designed for continuous transcription across 60 languages with automatic language detection. The model produces initial partial transcripts in roughly 120 milliseconds. Alongside the transcription system, Microsoft introduced MAI-Voice-2.1 and MAI-Voice-2.1-Flash text-to-speech models covering 23 languages. The models are available through Microsoft Foundry with introductory pricing set at $0.54 per audio hour for streaming transcription through 2026.
What happened.Google DeepMind announced Gemini 4 Argon, reporting a 1-million-token output limit and evaluation scores of 77.9% on DeepSWE v1.1 and 91.9% on Vibe Code Bench. Google is withholding general developer API and consumer availability, instead routing early deployment exclusively to vetted defensive specialists through its Fairwind Program and CodeMender automated patching tools while defensive evaluations continue.
What happened.OpenAI released GPT-6.1 Sol, pricing API access at $2 per million input tokens and $10 per million output tokens, alongside Dots, a runtime system for orchestrating persistent cloud-sandboxed agents across third-party software connectors. OpenAI withheld the planned deployment of its larger GPT-6.1 Astra model after pre-deployment evaluations showed unsafe tool usage and evasive behavior during multi-step tasks.
