Explore the complete news stream by source, topic and date.
MIT Technology Review’s EmTech Future conference delves into the five pillars for building scalable AI systems in a heavily regulated environment—data, engineering, platforms, business models, and governance—and underscores how trustworthy, responsible deployment can transform emerging capabilities such as agentic AI into sustainable investment value.
According to research published in Science, Microsoft’s AI diagnostic system MAI-DxO achieved an accuracy of 83% in simulating the diagnosis of complex cases, significantly outperforming general practitioners at 36%; however, analysis indicates that this performance improvement stems primarily from the advanced foundation models it leverages—such as GPT-5.5—while the marginal gains attributable to its proprietary DxO orchestration architecture are limited, and no clear advantages were observed in routine settings like emergency departments.
Mark Zuckerberg is vigorously advancing a vision of the future built on autonomous AI agents—Muse will shop, work, and even engage in conversations with other AI agents on behalf of users. This vision gained attention following Meta’s Q3 2026 earnings call; Muse was made available to the public last month, though its actual capabilities and large-scale deployment remain in the early stages.
Donating Agent Skills to the Agentic AI Foundation News today from Anthropic: We’re committed to keeping Agent Skills open, community-driven, and vendor-neutral. Today, we are donating the standard to the Agentic AI Foundation (AAIF) under the Linux Foundation, which we co-founded with Block and OpenAI last year, and to which we donated the Model Context Protocol (MCP). I'm pleased to note I called this one last December, when Anthropic first declared skills to be an "open standard": I wouldn't be surprised to see this end up in the AAIF, recently the new home of the MCP specification. Fun side note: since my blog content is available to Claude via the datasette-mcp plugin I found that post by prompting Claude to "Use simonwillison.net MCP to find if I ever talked about how I want skills t
Amazon Polly’s generative engine has undergone a major upgrade, significantly enhancing speech naturalness, accuracy, and emotional expressiveness, while supporting low-latency streaming synthesis. Fourteen new generative voices have been added, covering nine regions, and for the first time, generative voices in Hindi (Arjun, Kiara) and Finnish (Suvi) are universally available. The entire update is a seamless, drop-in replacement, requiring no changes to API calls or code.
Spanish pet-care company IskayPet leveraged Amazon Bedrock Managed Knowledge Bases to build an enterprise-grade knowledge assistant—covering Spain, Portugal, and France, and supporting four languages (Spanish, Portuguese, French, and English)—without needing to develop its own RAG infrastructure. The assistant supports business scenarios such as retail, veterinary care, onboarding, and procurement.
AWS has announced that SageMaker AI’s serverless model customization feature now includes “Training Sessions,” which are fully available starting today. Users can independently define post-training logic—such as custom loss functions, reward signals, or multi‑round reinforcement learning—while the platform automatically handles GPU allocation, distributed execution, fault tolerance, and checkpoint storage.
Amazon SageMaker has released the Training Sessions API, enabling users to implement custom large-model training workflows—such as SFT, RLHF, GRPO, and KTO—using pure Python scripts. GPU resource scheduling, fault tolerance, distributed execution, and checkpoint management are all handled by the service.
Microsoft AI CEO Mustafa Suleyman told The Verge that the company has developed an AI diagnostic tool whose accuracy in testing for multiple diseases is up to four times that of human doctors; the findings were published in the top-tier journal Science, based on a comparative evaluation of real-world medical cases and emergency‑room data against Microsoft’s proprietary benchmark (name not fully disclosed).
Amazon Quick is an HIPAA‑compliant service offered by AWS that enables healthcare organizations to process Protected Health Information (PHI) upon signing a Business Associate Agreement (BAA); however, compliance with HIPAA does not equate to full regulatory compliance. Customers must rigorously manage its PHI processing workflows, encryption mechanisms, logging policies, and irreversible configuration requirements.
AWS has released a technical solution that leverages the Kokoro-82M text-to-speech model hosted on SageMaker, using bidirectional WebSocket to enable AI agents to generate text and synthesize speech simultaneously, significantly reducing end-to-end speech response latency. The solution offers two audio back‑transmission methods—Base64‑SSE and unified binary transmission—and is explicitly labeled as “proof-of-value,” meaning it has not yet reached production‑ready status.
Elastic GPUs, sandbox growth, and the economics of the agent cloud
Elastic GPUs, sandbox growth, and the economics of the agent cloud
According to The Verge, well-known modder Sammy Hajhamid (online alias pepsipu) claims to have successfully jailbroken the Nintendo Switch 2. His tool, ‘ns2jb,’ was developed with assistance from OpenAI Codex and exploited a vulnerability triggered by an older game. A demonstration video shows “Mario Kart World” running a modified version that includes unofficial characters such as Cinnamoroll, the red Angry Birds, Peter Griffin, and Minecraft’s Steve.
Microsoft’s security team reported that a China-linked hacking group compromised the cloud backup service of IT management provider ConnectWise, stole customer application credentials, and leveraged the open-source AI framework ARTEX to carry out rapid lateral movement, launching nearly 3,000 login attempts per minute on average and resulting in espionage activities affecting approximately 2,250 client organizations.
The AWS blog introduces an empirical, automated scaling method for deploying text-to-speech (TTS) models on SageMaker real-time endpoints, eliminating guesswork and basing the approach on two measured metrics—Q_max (the maximum sustainable concurrent requests per instance) and T_total (the time required for planned scaling)—to create a reproducible, verifiable scaling strategy.
Rescale, in collaboration with AWS engineers and leveraging the AI-Driven Development Lifecycle (AI-DLC) methodology, the Strands Agents framework, and Amazon Bedrock (Claude Opus 4.8), completed within five days a production-ready AI test-routing agent designed to automatically classify and route nighttime end-to-end test failures, and rolled out this human-AI collaborative development paradigm across the organization.
Laika Animation Studio’s new film, “Wildwood,” took five years to produce and involved more than a thousand artists; the entire film was created using traditional stop-motion animation techniques. Director and CEO Travis Knight publicly stated that generative AI “runs counter to everything our studio stands for.”
Uber announced that it has rolled out ChatGPT Work to all roughly 30,000 of its global employees, scaling up from an initial pilot involving a thousand staff at the start of the year to company-wide adoption. The tool is already being used for data analysis, localization of multilingual training materials, and building interactive work dashboards, though specific use cases remain under each team’s own exploration, and the company has not yet disclosed the procurement costs.
Microsoft Research has unveiled Flint-interactive, a domain-specific language for building highly expressive, semantically rich interactive data visualizations. Its compiler automatically maps user interactions—such as clicks and brush selections—to semantic events and uses high-level operators to drive coordinated updates across multiple elements, including charts, tables, and text, significantly reducing the cost and development complexity of creating reliable interactive visualizations with AI agents.
