All news

UUnite.AI22:45

AI score48

LinearB Launches Agentic Studio to Provide Engineering Context for Coding Agents

LinearB has released Agentic Studio, which organizes software delivery data into context usable by engineering leaders and coding agents. The platform is now available to all LinearB customers. The first agent skill allocates effort based on task risk and complexity, and is in early access. LinearB is also exposing delivery metrics and bottleneck analysis through MCP, and says developers using AI merge 96% more pull requests per week than a year ago.

#Product updates#Agents#English text is being prepared.

SSiliconANGLE22:45

AI score39

LinearB Launches Agentic Studio to Provide Context for Human Engineers and AI Coding Agents

LinearB has released Agentic Studio, which consolidates data from Git repositories and project management tools into a unified context layer, enabling engineers to query release bottlenecks in natural language and, through MCP, providing coding agents such as Claude Code and Codex with context including rework rates, historical incidents, and code ownership. LinearB's report shows that AI-assisted pull requests are about 2.6 times the size of ordinary PRs, with review wait times more than 5 times longer; Kraken used its MCP integration to identify that review delays stemmed from time zone mismatches, cutting review time by more than half within four months. The first effort skill is now available in early access, and can dynamically adjust an agent's computational investment based on PR complexity.

#Product updates#Agents#English text is being prepared.

SSiliconANGLE22:45

AI score23

Guardz integrates Palo Alto Networks' Prisma Browser for…

The cybersecurity startup Guardz has integrated Palo Alto Networks' Prisma Browser for Business browser protection into its security platform for small-business MSPs, to block attacks in which phishing emails lead employees to fake login pages. Guardz correlates browser events with the original emails, Microsoft 365 activity and endpoint telemetry under the same user identity, and hands them to its 24/7 MDR team to handle as a single incident. Guardz handles deployment and default configuration, and has added per-customer AI usage policies and AI tool usage reporting.

#Product updates#Safety / alignment#Agents

SSiliconANGLE22:45

AI score21

MazeBolt Launches Native RADAR Deployment for Akamai DDoS Protection Customers

MazeBolt has launched native deployment of its RADAR validation platform for Akamai DDoS protection customers, enabling enterprises to start continuous testing within hours without taking services offline. RADAR launches simulated attacks from the customer's own environment, and in the first test of a production environment, approximately 37% of attack vectors were able to bypass existing defenses; this native Layer 7 support is now available to all RADAR customers.

#Product updates#Safety / alignment#English text is being prepared.

LLeikeji22:44

AI score40

Six AgiBot Genie G2 robots enter a Shanghai convenience store, taking on a 24-hour real-world work challenge

On November 2, six AgiBot Genie G2 humanoid robots entered an operating convenience store in Xuhui, Shanghai, working continuously for 24 hours as greeters, baristas, stock clerks, and cleaners, under live online supervision from viewers across the internet. The six robots were centrally coordinated by the embodied Agent framework Genie Master, connecting task judgment, action execution, and result feedback through the Embodied Execution Protocol (EEP), and storing both successful and failed task trajectories in long-term memory to feed back into action policy optimization. This was the final real-world test following the release of AgiBot's AI WAVE series of technologies.

Same story · 6 reports

#Industry news#Embodied AI#Agents

SShao Meng22:43

AI score15

Shao Meng questions that domestic AI sharing still uses ERNIE Bot for comparison

Shao Meng complained that in the AI tool comparison at the MBA orientation sharing session of a top 3 domestic university, ERNIE Bot still appeared. He believes the content has not been updated for several semesters, and the AI industry experts who shared may not have paid attention to AI for a long time. He pointed out that the level of this sharing session was not low, and the audience was basically corporate executives.

#English text is being prepared.

JJosh Woodward22:42

AI score3

Google executive Josh Woodward wishes Shah Rukh Khan a happy birthday

Shah Rukh Khan posted "Dost ho toh Google Gemini jaisa" and thanked fans for their love. Google's Josh Woodward reposted it, writing only "HBD SRK!" to wish him a happy birthday.

Quoted postShah Rukh Khan · @iamsrk

Dost ho toh Google Gemini jaisa.. Thank you for the KING-sized love always.... had to sneak out for a drive to get this photo ha ha.. @GeminiApp https://t.co/z3kLfG2Wf5

#English text is being prepared.

SServeTheHome22:42

AI score62

Qualcomm Completes Preliminary Validation of HBC Stacked Memory, Dragonfly AI250 to Begin Sampling in First Half of Next Year

Qualcomm announced that it has completed preliminary validation of its High Bandwidth Compute (HBC) stacked memory architecture, which stacks multiple layers of DRAM on a small compute die for use in its Dragonfly data center AI accelerators. The testing used early samples with 2 layers of DRAM, while production products will have 4 layers; Qualcomm claims that its energy efficiency per unit of bandwidth per watt exceeds the HBM4 specification by 5 times, and that measured bandwidth is 18 times that of external LPDDR5X, but it did not disclose specific bandwidth figures. The first Dragonfly AI250 integrating HBC will begin sampling in the first half of next year, and Qualcomm also plans to use this technology in Snapdragon consumer products.

#Industry news#English text is being prepared.#English text is being prepared.

TTech.eu22:42

AI score34

Roleo, founded by a Veriff co-founder, raises €2.9 million to expand its AI career agent across Europe

Roleo, founded by Veriff co-founder Janer Gorohhov, has completed a €2.9 million funding round led by Practica Capital with participation from Inventure, to expand its free personal AI career agent for tech professionals across Europe. The agent continuously monitors the job market, evaluating hundreds of thousands of opportunities each week based on users' experience, salary expectations and preferences, typically shortlisting the 3 most relevant positions for each person, and can connect them directly with hiring managers. Since early 2026, it has connected users with 30,000 career opportunities.

#Industry news#Agents

MMediaNama22:42

AI score24

MobiKwik Turns Profitable in Q2 FY2027 with Net Profit of Rs 1.18 Billion, 92% of Code Generated by AI

MobiKwik posted a net profit of Rs 1.18 billion in Q2 FY2027, turning around from a loss of Rs 286 million in the same period last year, with revenue up 9% year-on-year to Rs 2.957 billion. The company said AI now generates 92% of its code (up from 80% last quarter), Help LLM resolves 91% of customer queries, and it has achieved 100% AI-led fraud detection, aiming to become an AI-first company by FY2028.

#English text is being prepared.

99to5Mac22:42

AI score66

Wired Investigation: Over 400 AI Apps with Hidden Pornographic and Violent Deepfake Features Passed App Store Review

A Wired investigation found that a network suspected of being operated by Chinese entities includes more than 400 domains and apps. These apps passed review by presenting everyday features such as language teaching and AI fitness coaching, then loaded interfaces capable of generating pornographic and violent deepfake content under specific conditions. They drew traffic through ad placements on Meta's platforms, and some apps had been live for a year. After Wired provided details, Apple removed 19 apps, saying the developers had swapped the source content websites after listing; Meta said it had taken action against more than 10,000 ad accounts associated with these app domains.

Same story · 2 reports

#Safety / alignment#Image generation#Video

99to5Mac22:42

AI score40

Canva launches Affinity all-in-one design app for iPad, free for individual users

Canva has released Affinity for iPad, integrating sketching, vector, photo, and layout editing into a single app that individual users can download for free. The app is built specifically for touch and Apple Pencil, supporting Apple Pencil Pro's pressure, tilt, rotation, squeeze, and hover, and offers AI features such as Generative Expand, upscaling, and motion blur. The desktop version of Affinity has surpassed 7 million registered users in its first year since launch.

Same story · 3 reports

#Product updates#Image generation

99to5Mac22:42

AI score25

Report: MacBook Pro to See Multiple New Models Next Year

Bloomberg reports that Apple will launch multiple new MacBook Pro models next year, including a second-generation touchscreen MacBook Pro equipped with M7 Pro and M7 Max chips. Apple had previously planned to skip the M6 Pro and M6 Max and move directly to the M7 series, which is designed primarily around a major boost in on-device AI processing. Currently, the touchscreen version requires a $400 premium, and Gurman expects Apple will eventually merge the touchscreen and non-touchscreen high-end product lines.

#Industry news#English text is being prepared.

XX.PIN22:41Product release

AI score74

DeepSeek launches a team-level Super-agent that can automatically monitor group chats and fix bugs

DeepSeek has launched a new Super-agent that can monitor developer group chats and PRs, and automatically reproduce and fix bugs mentioned by team members without requiring an @ mention. The agent adopts event-driven triggering, synchronizes context across the entire team by decoupling storage and computation, and achieves continuous background operation with the help of an external Watchdog system; DeepSeek also uses it internally to review code and iterate on its own Harness codebase.

Same story · 24 reports

#Product updates#Agents#DeepSeek

MMeituan LongCat22:41Model releases

AI score73

Meituan LongCat Releases LongCat-2.5 Multimodal Agent Model with 1.6T Total Parameters and ~48B Activated

Meituan LongCat has released LongCat-2.5, its latest native multimodal agent model, with 1.6T total parameters and approximately 48B activated parameters. The model targets long-horizon agent tasks and enhances visual understanding and agentic coding capabilities, covering complex coding projects as well as visual coding work such as creative design and 3D. For more information, see https://t.co/ULLDweplqR.

Same story · 11 reports

#Model releases#Agents#Multimodal

MMacRumors Apple Intelligence22:41

AI score43

Canva Brings Free All-in-One Affinity App to iPad

Canva has released Affinity for iPad, integrating sketching, vector, photo editing, and layout into a single interface, and adding Sketch Studio built for touch and Apple Pencil. Users can switch between the Sketch, Vector, Pixel, and Layout studios within the same .af file, and files can also be opened on desktop; core tools are free for individuals, but a free Canva account registration is required, while AI tools require a paid Canva plan. The iPad version requires iPadOS 18.0 or later, and at launch lacks features such as Books, Batch, Data Merge, and macro recording.

Same story · 3 reports

#Product updates#Image generation#English text is being prepared.

CComputerworld22:40

AI score32

Windows 11 upgrades stall amid "chipflation" and "RAMageddon"

With CPU price hikes ("chipflation") and memory price hikes ("RAMageddon") driving up PC prices, Windows 11 upgrades have stalled, and Microsoft's Windows revenue from PC makers fell 27% in the most recent quarter. StatCounter data shows that in September 2026, Windows 10 still accounted for 27.7% of Windows PCs worldwide, while Windows 11's share fell from 72.57% in February to 71.5%. Analysts expect PC prices may not ease until 2028 or even 2029, and Counterpoint forecasts that PC shipments in the fourth quarter of 2026 will decline 27% year over year.

#English text is being prepared.#Industry news

FFigma AI / Agent / Make22:34First-party

AI score20

How Figma brought Hello Kitty and Friends into FigJam

Figma partnered with Sanrio to launch Hello Kitty and Friends in FigJam for one week starting November 1. Users can choose one of 12 Sanrio characters to join brainstorming, and each character comes with its own cursor, washi tape, and sticky note theme. The team used Figma Motion to create the loading animation and used Figma agent to complete batch edits; a free Hello Kitty birthday pop-up event will also be held in New York from November 6–8.

#Product updates#Agents

QQbitAI22:30Model releases

AI score65

SenseTime Releases Multimodal Agent Model SenseNova 6.8 Flash

SenseTime has released the multimodal agent model SenseNova 6.8 Flash, which focuses on delegated intelligence: users only provide a goal, and the AI handles the entire workflow. In official evaluations, it ranked first on TeamBench and DataAnalysis2.0, and tied for first with GPT 6 Luna on BEAM 128K. The author tested dynamic web pages, dynamic posters, SVG animations, and dynamic PPTs in the SenseTime Xiaohuanxiong client, and also tested long-term memory and proactive interaction; the model can proactively remind users of data corrections in old materials.

Same story · 5 reports

#Product updates#Agents#Multimodal

PPhoenix Technology22:30Product release

AI score66

Kimi Releases Game Development and Creative Work Solution, Agent Can Operate Unity, Unreal Engine, and Blender

Kimi has released a game development and creative work solution, based on the Kimi K3.1 Pro model, Kimi Code, and the Kimi Work series of Agent products, and integrated with professional tool plugins such as Unity, Unreal Engine, and Blender. The solution provides game teams with assistance capabilities ranging from gameplay prototype validation and engineering and art production to publishing and operations; the plugins can directly operate existing projects and deliver files and operation records that can continue to be edited. Kimi claims that in one demonstration, the model worked continuously for 20 hours without human intervention, producing a native project that can be sustainably developed.

Same story · 2 reports

#Product updates#Agents#English text is being prepared.