News
Updates, product notes and practical coverage for AI platforms and operators.
AWS has added Meta’s Muse Spark Open to SageMaker JumpStart in BF16 and NVFP4 variants. The open-weight coding model has 816 billion total parameters, activates 42 billion per token and supports a one-million-token context across text and multimodal inputs. Customers can deploy controlled endpoints, but multi-GPU cost, licensing and tool permissions require careful evaluation.
Meta is adding natural-language controls to Facebook Feed and Reels. A new Your algorithm area lets people manage preferred topics, while follow-up prompts after Interested or Not interested accept freeform feedback. Large language models interpret intent and context so recommendations can change within the same session.
Meta has introduced Muse Wonder and Muse Wonder Full, its most capable agent models to date. They target coding and expert-domain tasks, with Active Memory for carrying useful context across conversations and Active Focus for recovering from errors. Wonder is available through Meta Model API and Muse Code, with Full for Muse Code subscribers.
Meta has published a discussion with infrastructure chief Santosh Janardhan about why the company builds its own data centres for AI. The account highlights power planning, chip selection and custom infrastructure as strategic product inputs, underscoring how frontier AI competition increasingly depends on long-term physical capacity rather than software and models alone.
Meta has answered three common criticisms of data centres, focusing on cooling water, household energy costs and local employment. The company points to closed-loop liquid cooling, direct funding for required grid infrastructure and long-term construction and operations work. The claims add useful detail but still warrant site-specific, independently testable reporting.
Meta has launched Muse for Small Business, extending its personal AI agent with business context, connectors and proactive workflows. Owners can link common commerce, finance, productivity and social tools, while approval gates are intended to stop Muse from publishing, sending messages or spending money without consent.
Meta announced on 1 October that Ray-Ban Meta Gen 3 is available in India, with 27 frame-and-lens combinations, longer claimed battery life and a customisable action button. The release also previews Muse access on AI glasses, but that agent is not yet the same as today’s Gen 3 availability. We examine the practical hardware changes and rollout caveats.
Meta has announced an Enterprise Platform intended to bring its Muse agent, Meta Business Agent, Muse API and Muse Code to businesses and developers. Chirantan “CJ” Desai joins as chief enterprise platform officer. The announcement sets a direction rather than publishing general-availability dates or prices, so organisations should distinguish the strategic launch from products they can contract for today.
Meta has unveiled VR Glasses weighing about 100 grams, with a separate compute-and-battery puck, 5K display, hand and eye controls, and an operating-system AI agent. It plans US sales in spring 2027 at US$1,299.99. Media, work and hologram-call claims are ambitious; Australian price, availability and real-world comfort remain unconfirmed.
Meta has outlined new Muse agent capabilities from Connect 2026: more retail and work connectors, a voice mode that can continue tasks during conversation, a planned agent email address, and future access through AI glasses. The 24 September announcement describes mixed rollout states. Users should verify availability and authorisation before letting a personal agent transact or communicate on their behalf.
Meta has announced Ray-Ban Meta Audio, its first audio-focused AI glasses, alongside a wider frame range. The new glasses weigh 43 grams, promise up to 12 hours of use and let wearers speak to an AI assistant without a screen. US pre-orders start at US$349; regional availability and the forthcoming Muse integration need checking before purchase.
Meta has announced new software and availability for Ray-Ban Display glasses, including voice-driven navigation, an Explore app and a personalised hologram for video calls. The hologram uses a short capture in the Meta AI app and is slated for US WhatsApp Early Access later in the year. Several other features and new-market sales have separate rollout dates.
Meta has introduced Muse Realtime Avatar, a system that turns streaming speech into responsive video characters. Its research post describes synchronised voice and visual generation, a distilled two-step video model, subsecond response measurements and invisible watermarking. The demonstration is distinct from the earlier Muse agent launch, and Meta cautions that not every shown character is available in the app.
Meta has announced Meta One, bringing AI and social subscription benefits into a broader family of plans. Personal, creator and business options serve different needs, while eligibility and rollout remain important qualifications. Australian users should check the offer inside their own apps before assuming every advertised feature is available.
Meta has launched Muse, a personal agent that works in a dedicated cloud computer and can continue tasks in the background. The initial rollout is in the United States. Its separate permission system and activity views are central to evaluating the service, while stronger confidential-computing protections remain a future commitment.
Meta says it will sign the EU AI Act Code of Practice covering transparency for AI-generated content. The commitment aligns its labelling and detection work with a European framework, while leaving implementation details to evolving technical standards and further work with the EU AI Office.
Meta is rolling out a more agentic Meta AI powered by Muse Spark 1.1. The assistant can plan multi-step work, connect with email and calendars, create presentations, run recurring briefings and accept real-time direction while a task is underway.
Meta is expanding its Louisiana data centre to 5GW of compute capacity as part of a US$50 billion AI infrastructure investment. The project is also reshaping the local economy through jobs, supplier contracts, education funding and major infrastructure upgrades.