Search “AWS”

Pichai Reviews Gemini 4 Argon: Why 3.5 Pro Was Cut

Google CEO Pichai revealed in the "Sources" interview that Gemini 4 Argon processed about 25 trillion tokens in 72 hours of launch, 3 times that of Gemini 3.0 in the same period, and is still rate-limited due to compute constraints. He explained for the first time the reason for cutting Gemini 3.5 Pro: the team found a gap in the model on Agentic Coding, first used the Flash version to complete intensive iterations from 3.5 to 3.8 in about 3 months, while Argon pretraining had already been completed at that time, so they injected the accumulated gains into Argon post-training. He also mentioned that Google internally has entered the RSI stage of researchers and models collaborating, and said the next-generation Gemini 5.0 will be much stronger than Argon.

European companies urged to participate more actively in open source software development

At the Open Source Summit Europe hosted by the Linux Foundation in Prague, Thierry Carrez, General Manager of Linux Foundation Europe, and others called on European companies to not only use open source software but also contribute upstream and participate in open source governance, in order to reduce dependence on a single vendor or country. A Linux survey shows that more than 90% of European organizations regard digital sovereignty as a key priority in their technology strategy; companies with mature open source governance have an average return on investment of more than 4x, compared with 3.6x for those without a formal governance structure. GitHub data from October shows that currently one in every three pull requests involves an AI agent, compared with less than one in ten a year ago.

Railway suffers cyberattack leaking customer keys, asks users to change API keys

Cloud hosting platform Railway discovered abnormal activity on October 21 in the system storing customer environment variables, stating that an unauthorized third party may have accessed encrypted environment variables and decrypted some of them, and has advised all users to change credentials such as API keys. Railway believes the attack may be part of a recent series of attacks targeting platform companies with the aim of breaching AI inference service providers, and has notified upstream suppliers to revoke potentially leaked keys. The company also released security guidance for OpenAI, Anthropic and GitHub users, advising them to delete potentially exposed API keys, generate new keys and check usage records; on October 23 its builds and deployments had recovered, but new user registration remained suspended.

OpenAI Codex Releases Version 0.164.0

OpenAI Codex has released version 0.164.0, adding ten numbered /presets slots for saving reusable prompts and slash commands, and supporting inline /model and /permissions parameters to directly set the model, reasoning effort, speed, and permission configuration. This version also supports enterprise MCP connection authentication, credential masking for remote Bash and Zsh sessions, and enables the Amazon Bedrock setup wizard by default.

Claude Code v2.1.309 Released: Fixes Multiple Issues Including Sandbox and Permission Checks

Claude Code v2.1.309 has been released, primarily fixing an issue where sandboxed Bash commands on macOS and Linux executed shell startup files outside the sandbox before each command, as well as an issue where macOS sandbox denyWrite glob patterns did not take effect through symbolic links. This version also fixes issues where sandboxed commands and auto-accepted edits could modify the MCP configuration file passed via --mcp-config, and where disabled MCP servers were reconnected by subsequent permission prompts. In addition, many git commands run by Claude Code itself on Windows have been sped up.

How Mixx Technologies Paves the Way for All-Optical Scale-up Networks

Photonics startup Mixx Technologies is betting on all-optical scale-up networks. Its first product, the SxC connector, can accommodate 64 fibers within a single connector, and the company claims it can arrange 1,000 fibers within 800mm2. The company has laid out a three-step roadmap of NPO optical engines, CPO, and optical circuit switches. Its first NPO module is targeted for launch in 2028 or 2029, and CPO products are expected to be released by the end of this decade. If Nvidia's Vera Rubin NVL576 is estimated at one pair of fibers per SerDes, it would require approximately 20,736 pairs of fiber connections to connect 8 NVL72 racks.

Amazon SageMaker AI launches Training Sessions API, enabling custom training loops without managing GPUs

Amazon SageMaker AI has added a Training Sessions API to serverless model customization, turning operations such as forward_backward(), optim_step(), and sample() into API calls. The model stays loaded on the GPU, and users only need to write the training loop in a local Python process, while GPU allocation, distributed execution, fault recovery, and checkpoint storage are handled by the service. The API is billed by tokens processed, and steps such as local reward computation do not count toward GPU time. The accompanying cookbook provides SFT and RL recipes; in the example, the GRPO loop for GPT-OSS 20B trains for 10 steps on GSM8K, raising held-out set accuracy from 56% to 75%; the Code RL recipe raises test accuracy from 40% to 50.5% after 50 training steps.

Why AI compute is rectangular—a conversation with Modal CEO Erik Bernhardsson

Modal CEO Erik Bernhardsson explains that traditional compute contracts, signed for a fixed number of GPUs and a fixed duration, are rectangular, whereas real workloads fluctuate; Modal provides elastic slices by aggregating demand from thousands of customers. Modal has surpassed $1 billion in annualized revenue, completed a $750 million Series D at a $16 billion valuation, and has cumulatively launched nearly 3 billion sandboxes. Sandboxes currently account for about one-third of Modal's revenue, with growth driven mainly by coding agents, background agents, and reinforcement learning.

Sakana AI Releases Ontology-Based AI Agent Platform Sakana Aquarium

Sakana AI has released Sakana Aquarium, an AI agent building and operations platform that centers on ontology, organizing the knowledge, rules, and data scattered across an organization into a unified semantic layer so that agents can confirm information, make judgments, and execute tasks. The platform comprises three layers of mechanisms—ontology, governance, and managed agents—and allows access permissions and pre-execution approvals to be configured, while recording the entire process from the basis of a judgment to the execution result. The organized data and ontology can be directly reused by other agents, and Sakana AI says it will subsequently develop mechanisms that let agents continuously update the ontology as the organization changes, as well as AI-assisted agent building.

Google opens the infrastructure behind Gemini agents as a cloud service, Agent Compute Engine

Google announced that it will open a preview of Agent Compute Engine to Google Cloud users later this year, providing each AI agent with a hardware-isolated microVM that balances container startup speed with sandbox security. The service was originally internal infrastructure for Gemini, Antigravity, and Google's model training pipelines; when an agent is idle, it checkpoints the microVM to free up resources. Google is also offering an open-source runtime, Agent Substrate, which can run on any Kubernetes cluster; the company says that in a benchmark of 200,000 agents it achieved 500 activations per second, with suspend and resume taking under half a second.

Anthropic IPO Draft Prospectus Revealed: $16.3 Billion in Revenue, $80.7 Billion Net Loss in Half a Year

Anthropic's draft IPO prospectus shows $16.3 billion in revenue and a net loss of $80.7 billion in the first half of 2026, of which $72 billion comes from the revaluation of financing instruments such as convertible notes and is not actual cash spending. Computing and infrastructure spending in the first half of the year was nearly $9 billion, about 55% of revenue, down from more than 200% in the same period last year, and the company had already begun to be profitable on an adjusted basis in the third quarter. The article points out that the real risk lies on the cost side: signed forward computing and infrastructure commitments exceed $600 billion and could reach as high as $750 billion; Reuters said that of the $518 billion at the time, about 80% was irrevocable or required minimum payments, while of the $16.3 billion in revenue in the first half of 2026, more than $3.2 billion came from just two major customers, and many major customers had not signed long-term purchase commitments.

Browser Use Roundup of the Best Browser Agent MCP Servers of 2026

Browser Use has released its 2026 roundup of browser Agent MCP servers, dividing such servers into two modes—driver and delegate—and evaluating four of them one by one: Browser Use Cloud MCP, Browser Use open-source MCP, Hyperbrowser MCP, and agent-browser chat. In a comparative evaluation across 100 BU Bench V2.1 tasks, the author found that the four driver-type tools consumed a median of 2.1M to 6.7M input tokens per task, all of which count toward the coding Agent's context. The article also provides each server's installation commands, model selection, billing methods, and known limitations, and explains that long tasks, parallel tasks, and structured-output scenarios are better suited to the delegate mode.

Cognition Expands in London: Devin Customer Demand Drives New Office and Executive Team

Cognition has opened a new office in London, expanding its UK team by more than 200% in six months and bringing in executives including Narry Singh and Aman Thind. The Access Group has deployed Devin to more than 2,000 engineers across 250 product lines, with merged PR volume increasing fivefold and some legacy migration tasks accelerated by up to 480x. Cognition previously raised over $2 billion at a $48 billion valuation and released the coding model SWE-2.

Stanford HAI launches AI economic indicators dashboard to track AI's real-time impact on the labor market

The Stanford HAI Digital Economy Lab launched a set of monthly dashboards called AI Economic Indicators in June 2026 to track AI's economic footprint in real time. Data shows that employment in AI-exposed occupations such as software development and customer service has fallen by more than 4% this year, and employment of young workers aged 22 to 25 in these fields has declined by more than 3% annually since ChatGPT was released in November 2022, with the decline widening to 4.4% in recent months. The dashboard integrates statistics from private companies, academic institutions, and government data, and is divided into the Transformation Tracker, which monitors 12 macroeconomic indicators; the Canaries Dashboard, which provides early warnings based on ADP Research payroll data; and the Adoption Monitor, which uses surveys to track generative AI adoption rates.

Recursive releases laptop-class coding model Flint, says it was built mainly with AI involvement

AI startup Recursive has released its first model, Flint, a coding model that runs offline on a laptop and is freely downloadable on Hugging Face under the Apache 2.0 license; the company says its development process was mainly carried out with the involvement of its in-house Eureka Machine. Flint is an MoE model that activates only 2.5B of its 40B parameters at a time, generates at about 131 tokens/sec on a MacBook Pro with an M4 Pro, and outperforms other models under 50B parameters on SWE-Bench Verified. Recursive says Flint requires more guidance than large paid coding assistants and recommends breaking large tasks into small steps.

How U.S. Hunger Uses Amazon Quick to Connect Leadership with Data

The U.S. nonprofit organization U.S. Hunger uses Amazon Quick to let leadership directly explore organizational data and share KPI definitions and financial documents through Quick Spaces; 17 colleagues currently have access. Its data comes from its own platform Impact Manager, refreshed nightly at 2 AM ET from SQL Server into the in-memory engine SPICE. The organization has delivered more than 250 million meals in total, growing about 500% over four years while staff headcount grew only 15–20%.

English translation is being prepared.

How Acuity Health Transformed Patient Services with LiveKit Voice Agents

Acuity Health built a custom voice agent on LiveKit: 80% of incoming calls are fully handled by the AI agent, most appointments are completed within 1 minute, and 81% of appointment calls end with "thank you." Its agent typically runs on a Gemma 3 27B model optimized on LiveKit Inference, paired with turn-detection models tuned for different patient populations, supports voice and text, and connects to call center tools, task management software, analytics providers, and EMR systems. The team had previously been constrained by low-code platforms; after switching to LiveKit, they built a working prototype within a week, reached a development environment within a month, and use LiveKit Simulations to stress-test scenarios before every change goes live.

English translation is being prepared.

OpenAI says its model helped discover a critical PHP security vulnerability

OpenAI stated that its model helped researchers discover a critical security vulnerability in PHP under the Patch the Planet initiative; PHP is estimated to be used to run about 70% of websites. Over the past two months, OpenAI led an alliance of maintainers, researchers, and security vendors to prepare fixes and protections before the vulnerability was disclosed, because remediation was complex and required significant changes to the PHP engine. OpenAI said all sites using PHP can now upgrade versions to stay secure, and considers this one of the most impactful vulnerabilities in recent times and a new phase in which AI can discover security vulnerabilities at internet scale.