theterminal.space

Today · Fri 20 Nov

Qwen3.8-27B: 6.9M downloads, but Artificial Analysis says slow

Qwen3.8-27B Has 6.9 Million Downloads on Hugging Face. Artificial Analysis Rates It Slow and Verbose

AI-generated summary Alibaba released the Qwen3.8-27B open-weight model on Hugging Face, where it accumulated over six million downloads. Artificial Analysis rates the model slow and verbose compared with similar open models.

Earlier:

Waymo Banned From Every California Freeway Since Nov. 11

White Waymo Jaguar robotaxi with roof sensors on a San Francisco street with the city skyline behind

AI-generated summary The California Department of Motor Vehicles restricted Waymo from operating or testing driverless vehicles on all state freeways effective November 11, 2026. Waymo says it is working with the California Department of Motor Vehicles and other state officials diligently and responsibly to resume freeway operations.

Also covered byThe Robot ReportNBC NewsThe Terminal

Nvidia Singapore revenue: 20% billing share vs $104B entity

Nvidia's Singapore Line Was 20% of Billing in Its Last Filing, Against $104 Billion at the Entity

AI-generated summary Nvidia reported that Singapore accounted for 20% of customer billing in its filing for the quarter ended April 27, 2025, and later reached 22% in the quarter ended July 27, 2025. Nvidia says customers use Singapore to centralize invoicing while products are almost always shipped elsewhere, and reporting so far has found no evidence that Nvidia broke export rules.

Also covered bySıfırıncı DakikaCrypto Briefing
Earlier:

Wednesday 18 November

Introducing GPT-6.2 Astra and Sol

AI-generated summary OpenAI says it has introduced GPT-6.2 Astra and GPT-6.2 Sol as new models designed for professional work and everyday tasks. OpenAI says GPT-6.2 Astra delivers improved performance on complex professional assignments while GPT-6.2 Sol offers efficient capabilities for everyday use.

Also covered byRSS.bz9to5MacThe Terminal
Earlier:

Corma Raises $125M to Launch Defensive AI Foundation Models

AI-generated summary Corma raised $125 million in a Series A funding round to launch defensive AI foundation models designed specifically for cybersecurity. The funding round was led by NEA with backing from Sequoia Capital, Khosla Ventures, and Coatue, bringing the total raised to $185 million.

Also covered bySecurityWeekVenture HeraldThe Terminal
Earlier:

Genesis AI raises $500 million Series A to scale general‑purpose robots

Source-provided image accompanying Genesis AI raises $500 million Series A to scale general‑purpose robots

AI-generated summary Genesis AI secured $500 million in Series A financing, valuing the company at $3.5 billion post-money. The funding round was co-led by Premji Invest, Khosla Ventures and Eclipse with participation from NVIDIA, AMD Ventures, Zoom, Samsung Next and Ericsson.

Also covered byBloombergSiftedThe Terminal

Tuesday 17 November

Tencent releases Hy4 flash with 12 billion active parameters and lower task costs

Tencent headquarters towers under construction beside a prominent bus stop sign in Shenzhen

AI-generated summary Tencent released Hy4 flash on 16 November 2026 with 12 billion active parameters and a 1-million-token context window for routine workplace tasks. Tencent says the model is available on platforms including Yuanbao and WorkBuddy, with free access for WorkBuddy continuing through 30 November.

Also covered byFresh WeightsRSS.bzThe Terminal
Earlier:

Anthropic Cuts Its Bug Disclosure Window to 30 Days, and Google Project Zero Follows: What Maintainers Should Do

AI-generated summary Anthropic says it will reduce its coordinated vulnerability disclosure timeline for high- and critical-severity bugs to 30 days starting on 1 January 2027. Google Project Zero announced a plan to transition its vulnerability disclosure schedule down to 30 days by early 2028.

Also covered byAzeem HassniSI Agent NewsThe Terminal
Earlier:

GTIG exploit_agent: automated patch diffing and exploitation

Google Finds exploit_agent Diffing Patches and Firing Exploits With No Operator in the Loop

AI-generated summary Google Threat Intelligence Group reported an attacker-run framework named exploit_agent that compares web application code before and after a patch announcement, drafts an exploit for the difference, and runs it without human intervention. According to the report, the framework targets n-day vulnerabilities using public patches, and victim telemetry suggested the operator focused on French media, entertainment, and education websites.

Also covered byCloudNinjasamartuvshin.comajayk.sh

Bharat Semi Raises $24M to Build AI-Integrated Image Sensors

AI-generated summary Bharat Semi secured $24 million in an all-equity seed funding round led by Info Edge Group, South Park Commons, and Indusbridge Ventures. The New Delhi semiconductor startup will use the capital to develop AI-enabled programmable image sensors and expand its engineering team.

Also covered byThe TerminalChip Dispatch
Earlier:

How the Claude Code team automates software verification

SI Agent News. SI is the new AI. Two AI podcast hosts on a newspaper front page

AI-generated summary The Claude Code team states that roughly 90% of their software is written and reviewed by Claude. The team uses verification loops to delegate work to AI agents and maintain high code quality.

Also covered bynews.lavx.huBytechapThe Terminal
Earlier:

NVIDIA SkillEvaluator: Open-Source Tool to Test if Your AI Agent Skills Really Work (GitHub)

AI-generated summary NVIDIA SkillEvaluator is an open-source Python tool released on GitHub under the Apache 2.0 licence to evaluate AI agent skills and plugins. The software features three independent tiers for validation, deduplication and live evaluation, supporting systems such as Claude Code, Codex, and Copilot CLI.

Also covered byblockchain.newsThe Terminal
Earlier:

Monday 16 November

Inkling 2 Small weights: Apache 2.0, 500GB VRAM needed

Inkling 2 Small Weights Land on Hugging Face Under Apache 2.0, and the Card Asks for 500GB of VRAM

AI-generated summary Thinking Machines Lab made the weights of Inkling 2 Small downloadable on Hugging Face under Apache 2.0 on Nov. 12. According to the model card, the main checkpoint needs at least 500GB of aggregate GPU memory.

Earlier:

Fireworks AI breach: Uber, Cursor and Notion hit

Fireworks AI Breach Reaches Uber, Cursor and Notion, With 149,346 People Notified

AI-generated summary Uber is notifying 149,346 people that their personal information was taken in a security breach involving its inference provider, Fireworks AI. Cursor and Notion have also confirmed that their data was stolen in the same intrusion affecting Fireworks AI.

Earlier: