Devlery
BlogSkillsTags

Blog

Notes and analysis on AI development.

All [496]AI [495]AI Agent [443]AI Infrastructure [311]Developer Tools [305]AI Coding [188]LLM [135]Security [109]AI News [52]MCP [33]Open Source [29]AI Safety [23]Productivity [10]Enterprise [7]Gemini [6]Microsoft [6]Anthropic [4]DevOps [4]AI Search [3]Claude Code [3]GitHub [3]OpenAI [3]ChatGPT [2]Data [2]AI Ethics [1]AI Governance [1]Android [1]AWS [1]Education [1]Fintech [1]Google [1]Meta [1]Policy [1]Robotics [1]Vercel [1]Workflow [1]
Android is becoming an AI OS, and Gemini’s real gate is the platform

Android is becoming an AI OS, and Gemini’s real gate is the platform

Google Gemini Intelligence tries to turn Android from an app-launching OS into an intelligence system that can read context and act.

Meta AI incognito mode turns privacy into infrastructure

Meta AI incognito mode turns privacy into infrastructure

Meta Incognito Chat runs AI conversations through WhatsApp Private Processing. The key shift is not deleted history, but verifiable private inference.

SANA-WM 2.6B asks what a one-minute world model really costs

SANA-WM 2.6B asks what a one-minute world model really costs

NVIDIA SANA-WM claims 720p, 60-second world modeling from a 2.6B backbone. The real story is not video polish but the cost structure of open models.

After DAU comes DAA, why Baidu wants an agent metric

After DAU comes DAA, why Baidu wants an agent metric

Baidu proposed Daily Active Agents as a core AI-era metric. The useful question is not token volume, but how many agents actually complete work.

ChatGPT can read bank data, and finance AI starts with read-only trust

ChatGPT can read bank data, and finance AI starts with read-only trust

OpenAI brought personal finance into ChatGPT. The real story is not budgeting, but how AI products handle permission boundaries around sensitive money data.

Google Saw the First AI Zero-Day, and Security Timelines Are Changing

Google Saw the First AI Zero-Day, and Security Timelines Are Changing

Google GTIG disclosed the first zero-day exploit attempt it assesses was developed with AI, shifting how defenders should think about discovery and weaponization speed.

Codex Windows sandbox sets the baseline for local agent security

Codex Windows sandbox sets the baseline for local agent security

OpenAI disclosed the Codex Windows sandbox design, moving local coding agent security from app isolation to OS-level execution boundaries.

Superpowers turns coding agents into a process layer

Superpowers turns coding agents into a process layer

Superpowers shows how the coding-agent race is shifting from model quality alone to repeatable skills, TDD, reviews, worktrees, and verification.

SocialReasoning-Bench shifts agent evaluation toward duty of care

SocialReasoning-Bench shifts agent evaluation toward duty of care

Microsoft Research released SocialReasoning-Bench, arguing that agent evals must measure whether agents represent user interests, not only whether tasks finish.

China’s AI Agent Guidelines Point to an Intelligent Internet

China’s AI Agent Guidelines Point to an Intelligent Internet

China’s new AI agent guidance moves beyond chatbot rules toward registration, identity, interoperability, permissions, traceability, and an intelligent internet.

WaveSpeed’s 260-model LLM API moves model choice into the routing layer

WaveSpeed’s 260-model LLM API moves model choice into the routing layer

WaveSpeed now exposes GPT, Claude, Gemini and 260+ LLMs through one OpenAI-compatible API. Here is what that means for multimodal agents, routing, cost, and trust boundaries.

Agent Room gives coding agents a small room to wake each other

Agent Room gives coding agents a small room to wake each other

Agent Room combines MCP rooms and CLI stop hooks so Claude Code, Cursor, Codex, and Gemini can collaborate asynchronously around one shared transcript.