SuperMe has raised $20 million to build a network of AI profiles based on real-life experts. The platform selects appropriate experts to answer user questions and creates a group chat with their AI profiles.
Why it matters: Provides insight into a new AI-powered expert network platform that has secured significant funding.
Google's Gemini app now lets users generate, preview, and edit documents, PDFs, presentations, and spreadsheets directly within the interface. A new full-screen editor enables refinement of files without leaving the app. This update expands Gemini's utility for document-centric workflows.
AWS announced that NVIDIA Nemotron 3.5 Super VL, a 120B-parameter MoE multimodal model, is now available on AWS and detailed its deployment on Amazon EKS, EKS Auto Mode, and SageMaker HyperPod EKS. The model activates 12B parameters per token and supports text, image, document, and video understanding with a 1M-token context. The post provides architecture guidance, validation using NCCL benchmarks, and performance evaluation with NVIDIA aiperf, including concurrency sweeps to plan capacity. It also highlights integration with agent frameworks and NVIDIA Dynamo for serving.
Google Gemini Notebook now includes hundreds of textbooks from OpenStax and S Chand. Users can use these textbooks to generate custom study guides and quizzes. The update targets students seeking AI-assisted learning materials.
Elorian launched Glimpse, its first visual reasoning model, and co-released the DecoyCount benchmark with Mercor to test whether models genuinely count objects or rely on shortcuts. Frontier models scored below 50% accuracy on DecoyCount, with GPT-6 Astra highest at 43.17%, while Elorian reports Glimpse scored 50.48% in an internal single-run test. CEO Andrew Dai explains Glimpse uses visual thinking, generating and editing images as part of reasoning, distinguishing it from world models like Atlas and Genie 4.
The New York Times is testing an AI feature in its Cooking app that helps users modify recipes. It can recommend ingredient and equipment substitutions, scale recipes to different serving sizes, and adapt recipes to a cook's needs. The feature has sparked discussion among users. An example showed the tool advising that baby spinach should be added later than kale because it wilts faster.
Mistral has released the weights for its trillion-parameter AI model, specifically targeting cybersecurity applications. The company claims this open approach gives organizations more control over their cyber defenses compared to closed models. The release emphasizes the strategic value of open-weight models in security contexts.
AssemblyAI released AssemblyAI-TTS-1, its first text-to-speech model, now generally available as a hosted API for live voice agents. It is priced at $15 per 1M characters with unlimited concurrency, supports six languages, 4,000+ voices and instant voice cloning, and exposes HTTP, WebSocket and OpenAI-compatible interfaces with LiveKit and Pipecat reference agents. AssemblyAI's self-reported eval claims 93.7% entity accuracy and 47 ms median time to first byte, while Cekura's independent benchmark measured 159 ms median streaming latency and 81.5% reading accuracy. The launch lacks SSML, speed control, pronunciation dictionaries and emotion tags.
Why it matters: Readers evaluating TTS vendors for voice agents get concrete pricing, interface and benchmark details alongside independent latency results that qualify the vendor's claims.
Akai is giving early access to modders for its upcoming MPC Sample Key, which features open-source firmware. The device is similar to the MPC Sample but with keys instead of pads, and Akai is embracing the hacker community's vibe-coded modifications. This move signals a shift toward open-source hardware and community-driven innovation in music production gear.
Why it matters: Highlights a concrete product announcement where a major hardware maker adopts open-source firmware and vibe coding, offering readers insight into emerging trends in music tech and AI-assisted development.
Google has released Gemini 4 Carbon, its latest frontier model, claiming major gains in coding, agentic tasks, and cybersecurity. The model tops the Artificial Analysis Intelligence Index v5.0, outperforming Claude Opus 5.5 and GPT-6 Astra. Initially available only to select cybersecurity defense companies, it will later expand to partners in Google's Fairwind Program. Gemini 4 Carbon builds on the Argon model released earlier this month.
Why it matters: Provides concrete details on a new frontier model launch with benchmark rankings and a specific cybersecurity focus, offering timely industry insight.
Anthropic has introduced new tools designed to give companies visibility into how their spending on its AI models translates into usage and value. The move comes ahead of a planned IPO, aiming to increase transparency for enterprise customers. The tools help businesses understand what their AI investments are delivering.
This AWS blog post describes new capabilities in QA Studio, a reference solution built on Amazon Nova Act for agentic QA automation. The update introduces applications to organize tests and report health metrics, plus team-based user management with roles and cross-team sharing. Health metrics include pass rate, flaky-adjusted health scores, and trend charts via CloudWatch. Access control uses Cognito groups and a ceiling rule for shared applications.
Google DeepMind has introduced Gemini 4 Carbon, its strongest model to date, designed for long-horizon agentic coding and cybersecurity defense. It ranks #1 on the Artificial Analysis Intelligence Index v5.0 with a score of 59, leading 15 of 17 published benchmarks against rivals like GPT-6 Astra and Claude Opus 5.5. Access is initially limited to vetted cyber defenders through a pre-release testing process, with broader availability later. The model shows notable gains in vulnerability finding, exploitation, and patching, though some rival benchmarks remain competitive.
Why it matters: This is a concrete, recent frontier model release with detailed benchmark data and a clear cybersecurity focus, offering actionable insights for developers and security professionals.
Google's AI Mode returns indirect responses when asked who won the 2020 and 2024 US presidential elections, directing users to official sources like the FEC instead of directly naming the winner. In contrast, it provides direct answers for elections from 2000 to 2016. The behavior was observed in testing after social media complaints about cagey answers.
MiniMax has open-sourced MCode Stack, a collection of 17 SKILL.md agent skills and 5 checklists used by its MiniMax Code team, released under the MIT license on October 28, 2026. The stack focuses on harness engineering, including planning, context management, code review, and long-session handling, and installs into MiniMax Code or Claude Code. No benchmarks or eval results are published, so it serves as a practical playbook rather than proven performance. The skills are plain Markdown and compatible with agents supporting the SKILL.md format.
Google is launching a developer platform that allows businesses to integrate their services into Google's AI experiences, including Gemini and Search. This move could change how consumers discover and use apps as AI agents reshape the web. The platform aims to expand the reach of AI-driven interactions beyond Google's own services.
Chrome's AI-powered Autofill update can now pull flight confirmations, tracking info, order details, and other information from a user's Gmail inbox to fill out forms, with user permission. This expands the feature's capability beyond basic form data to leverage email content for more comprehensive autofill. The update highlights a growing trend of AI integration into everyday browser functions.
Chrome's enhanced autofill feature is expanding to use Gemini with Personal Intelligence to suggest information from Gmail, such as tracking numbers, flight bookings, and hotel reservations. This aims to save users from manually searching for confirmation emails or receipts when filling out forms. The feature builds on previous autofill improvements that sourced data from Google Accounts and Google Wallet.
