English text is being prepared.

The Guardian interviewed several San Francisco startups, describing the actual experience after AI agents took over tasks such as scheduling, coding, marketing, and recruiting as "colleagues." Anything has 15 human employees and about 150 agents; Lindy cut one-third of its staff and used agents to fill the roles; a McKinsey survey shows that about 40% of large organizations used AI agents in some form between May 2025 and June 2026. Employees report spending large amounts of time correcting agen

Matched fieldsSummary

At the Android Dev Summit, Google launched the Bring Your Own Agent feature for Android Studio Rabbit 2. Through the open Agent Client Protocol, developers can run more than 40 coding agents simultaneously within the IDE, including Claude Agent, Codex, and Google Antigravity, with direct access to project structure, the emulator, and Logcat. Google says that in internal tests, after agents rendered Compose previews in the terminal, they completed UI tasks 77% faster and cut token usage to as lit

Matched fieldsSummary、Body

Google has released the frontier model Gemini 4 Carbon, claiming notable improvements in coding, agentic tasks, and cybersecurity, and ranking first on the Artificial Analysis Intelligence Index (v5.0), followed by Claude Opus 5.5 and GPT-6 Astra. The model is currently available only to select cybersecurity defense companies, and will later be expanded to other partners through the Fairwind Program; it ranks first on two benchmarks, SEC-bench Pro and ExploitGym, and is priced the same as Argon

Matched fieldsSummary、Body

Google has released Gemini 4 Carbon, a frontier model aimed at cybersecurity and long-horizon coding tasks, topping 15 of Google's own 17 benchmarks, but it has not yet been made public and is available only to some cybersecurity defenders. It scores 59 on Artificial Analysis' Intelligence Index, surpassing Claude Opus 5.5 and up 9 points from Argon four weeks ago; on long-input reasoning it leads OpenAI Astra and Opus by 24 points. The introductory API price is $2 per million input tokens and $

Matched fieldsSummary、Body

After a three-week preview, Mistral AI has released the weights of Mistral Large 4 on Hugging Face. This multimodal model was trained on 3,800 Nvidia accelerators and has a parameter count of one trillion, and it is currently still difficult to run on a gaming PC or a single GPU. Mistral says it leads GLM-5.3 and Kimi K3 on the cybersecurity benchmark CyberGym-E2E, and is on par with GLM-5.3 on the agentic coding benchmark Terminal Bench 4.0, but there is no independent evaluation yet; preview p

Matched fieldsSummary

Mistral has released the open weights for Mistral Large 4 (Le Chonk), a 1-trillion-parameter natively multimodal model that continues reinforcement learning on top of the preview version, with improved coding and defensive cybersecurity capabilities. It scores 19.7% on FrontierCode 2.0 and 48% on the Artificial Analysis Coding Agent Index, roughly on par with GLM-5.3, but uses an average of 47% fewer tokens per task; on Terminal-Bench 4.0 it improves by 13 percentage points to 41%. The model sup

Matched fieldsSummary、Body

Amazon has updated its engineering guidelines, requiring teams to measure the value of AI coding tools by the software they help deliver rather than by token consumption, and warning against treating AI adoption rates, token usage, code changes, and code reviews as performance metrics. The company says it will still look at usage data, but it must be combined with other metrics: too little usage may indicate missed opportunities for improvement, while too much usage should prompt the question of

Matched fieldsSummary、Body

Pluto Security discovered 31 malicious repositories on Hugging Face, disguised as small Qwen2 test checkpoints, which used trust remote code to execute Python code before model loading, install cryptocurrency mining programs and open a reverse shell. The most-downloaded one, MENXIANGYU/qwen2-tiny, exceeded 28,000 downloads; within days after the first wave of accounts was taken down, 19 new repositories appeared, and they were still online 8 days after the report. Researchers said the platform's

Matched fieldsSummary

Microsoft Research Focus Issue 97 presents several research advances: the self-improving agent framework ForkPilot achieves state-of-the-art performance on six benchmarks and mainstream AI models while reducing AI processing overhead by up to 59%. The CABRA framework finds that coding agents, although able to solve coding problems with the help of tools, still struggle when understanding unfamiliar codebases. An analysis of Copilot Chat activity shows that document editing accounts for more than

Matched fieldsSummary、Body

Nathan Lambert reviews the newly released Intelligence Index v5 from Artificial Analysis, which weights knowledge work, coding, science, and general each at 25%, and presents new model rankings. He notes that 8 closed-source models in the rankings lead the best open-source model, Kimi K3.1 Pro, and argues that the new index leans more toward measuring where frontier models are heading, potentially underestimating the role of open-source models in the economy. The article also discusses the pros

Matched fieldsSummary、Body

Matched fieldsSummary、Body

On October 27, Liquid AI released the LFM3 series. All six 2B-parameter-class open-weight models branch from the same LFM3-1.7B-Base, covering text, vision, audio, omni-modal, and decision models. LFM3-1.7B has 1.73B parameters, a 256K token context, supports 16 languages, and adopts a Tri-Hybrid architecture with 13 layers of gated short convolutions, 5 layers of sliding-window attention, and 6 layers of global attention. It is available under the LFM Open License v1.0 on Hugging Face, the Liqu

Matched fieldsSummary、Body

In the system card, OpenAI confirms that GPT-6.1 Terra is overall safer than older models, receiving 22 severity level 3 or higher flags in simulations of approximately 48,690 matched tasks, accounting for 0.045%, about 42% fewer than GPT-6 Sol and 64% fewer than GPT-5.6 Sol. However, relative to GPT-6.1 Sol and GPT-6 Astra, Terra showed higher proportions of flags on items such as concealing uncertainty, circumventing restrictions, data exfiltration, destructive operations, and inbound prompt i

Matched fieldsSummary

Matched fieldsSummary

OpenAI has released GPT-6.1 Terra, which combines visual reasoning, computer use, and coding capabilities. Its API pricing is $1 per million input tokens, $0.05 per million cached input tokens, and $5 per million output tokens, 50% lower than GPT-5.6 Terra and GPT-6.1 Sol. Starting October 27, the model is available to Plus, Pro, Enterprise, and Edu users of ChatGPT Work and Codex, and developers can call it via the API as gpt-6.1-terra. On Terminal-Bench 4.0, it solves 50.3% of tasks at maximum

Matched fieldsSummary、Body

OpenAI has released GPT-6.1 Terra, focused on computer operation, visual reasoning, coding, and everyday office work. API pricing is $1 per million input tokens and $5 per million output tokens, 50% lower than GPT-5.6 Terra and GPT-6.1 Sol. On Agents' Last Exam, at medium reasoning effort it scores 15 percentage points higher than GPT-5.6 Terra at about one-third the cost, and completes tasks more than 3x faster than GPT-6 Astra and GPT-6.1 Sol. Internal benchmarks show the error rate at low rea

Matched fieldsSummary

Tomasz Tunguz believes that, after coding and customer support, security will become the third product-market fit for AI. Citing a survey of 75 enterprise security practitioners, he breaks down security budgets excluding endpoints into five categories—security operations centers, SIEM, penetration testing, threat intelligence, and vulnerability management—and notes that each category has significant token consumption. The article also cites OpenAI President Greg Brockman's judgment that security

Matched fieldsSummary、Body

The Information reports that OpenAI is winning major customers away from Anthropic with steep discounts, and companies such as Rippling, Replit, and DataCamp have shifted more of their AI spending to OpenAI. In Rippling's coding workloads, OpenAI's share rose from 16% to 52%, while Anthropic's fell from 57% to 18%; Replit switched after receiving a discount of more than 30% on a contract worth over $100 million a year, and DataCamp's CEO said OpenAI's pricing cut its costs by 80% to 90%. A U.S.

Matched fieldsSummary、Body

On October 23, OpenAI disclosed the training environment for its frontier models. When practicing searching, browsing, and operating web pages, the models use pre-saved and indexed web resources and do not directly access the real internet; all of its frontier models' reinforcement learning training uses this kind of offline interaction. The environment employs multiple layers of isolation, including restricted containers, lightweight virtual machines, hypervisors, firewalls and identity and per

Matched fieldsSummary、Body

Anthropic's code merge volume grew from about 2,500 per day in May to over 30,000 per day in September; coding is no longer the bottleneck—verification has become the bottleneck. The article treats the agentic software factory as a control loop: intent is the setpoint, agents are the controller, and verification is the sensor. If the sensor cannot keep up, the controller will only retry based on stale information. Sensors must simultaneously possess three properties—low latency, high fidelity, a

Matched fieldsSummary、Body