CÈ
Thursday, 27 August 2026
Get Intel

Index Company

Microsoft

58 CÈ pieces since 1 May 2026: 1 deal, 50 intel items, 7 essays. Newest first. All companies →

  1. Intel

    The model was never the moat: DeepSeek open-sources its agent harness under MIT

    DeepSeek published Harness v0.1 under the MIT licence, built on the Cordis meta-framework where models, tools, sandboxes, sessions, and orchestration are all plugins. It ships OS-level sandboxing and supports Anthropic, OpenAI, Bedrock, Azure, and Gemini.
  2. Intel

    Scanners that cry wolf are worthless: Strix proves every bug and passes 45,000 stars

    The open-source pentest agent runs the code, chains exploits, and attaches a working proof of concept and a fix patch to every finding. It has passed 45,000 GitHub stars and adds roughly 4,000 a week.
  3. Intel

    Defence gets cheaper than attack: Microsoft's 5B-active cyber model clears 95.95% on CyberGym

    MAI-Cyber-1-Flash activates 5B of 137B parameters and handles 90% of Microsoft's MDASH pipeline, escalating the hardest tenth to GPT-5.4. The system's CyberGym score rose from 88.45% in May at half the previous cost.
  4. Intel

    A breach turned AI security into shared plumbing: NVIDIA, Microsoft and IBM open-source their guardrails

    The Open Secure AI Alliance launched with more than 60 partners. NVIDIA put its NOOA agent-safety framework on GitHub, Hugging Face offered Safetensors to the PyTorch Foundation, and Microsoft contributed MDASH, its agentic bug-hunting harness.
  5. Intel

    Docs were written for humans: Meta's Astryx ships an agent-readable manifest, 9,000 stars in three weeks

    Meta open-sourced Astryx, the React design system behind Facebook, Instagram, and Threads, under an MIT licence. Its CLI manifest exposes every command, flag, and response type as JSON, giving coding agents a specification instead of prose documentation.
  6. Intel

    Who signs an agent's work when it writes the code? Block open-sources Buzz, giving each AI agent a cryptographic identity

    Block released Buzz, a free open-source workspace built on the Nostr protocol that gives every AI agent its own cryptographic identity and a signed chain of custody, positioning it against Slack and GitHub for agent-heavy engineering teams.
  7. Intel

    Security testing turns into a commodity agent: open-source Strix crosses 41,000 GitHub stars under Apache 2.0

    Strix, an open-source multi-agent penetration-testing tool that finds, validates, and exploits vulnerabilities across the OWASP Top 10, crossed 41,000 GitHub stars under an Apache 2.0 license, among the faster-climbing security projects of 2026.
  8. Deal

    Encrypted data is already being harvested for later: Singapore's pQCee raises $3.9M to make it quantum-proof

    The seed round, co-led by SGInnovate and Lotus One, follows $2.8M raised in 2022 and funds integrations already running with Thales, Microsoft and PQShield.
  9. Intel

    Agent orchestration turns into a public good: ByteDance's DeerFlow passes 77,000 stars under MIT

    DeerFlow, ByteDance's open long-horizon agent harness, shipped a ground-up 2.0 rewrite in late June. It orchestrates sub-agents, sandboxes, memory and extensible skills on LangGraph, and now carries 77,300 GitHub stars under an MIT licence.
  10. Intel

    When the build gets ten times faster, the tooling tax collapses: TypeScript 7 ships its Go rewrite, 11.9x quicker on VS Code

    Microsoft shipped TypeScript 7, a full rewrite of the compiler in Go, generally available July 8. Full builds run roughly eight to twelve times faster, with VS Code's own codebase compiling in 10.6 seconds instead of 125.7.
  11. Essay

    The biggest customer the AI labs have is quietly building their replacement

    Microsoft began routing tens of thousands of Excel and Outlook prompts a week to its in-house MAI models this month. The question that leaves for anyone building on AI is not which lab to trust.
  12. Intel

    OpenClaw's local-first AI assistant framework crosses 380,000 GitHub stars

    The self-hosted framework routes a personal AI agent across WhatsApp, Telegram, Slack, and Discord from a local gateway, reaching one of open source's fastest star counts on a privacy-first, cloud-independent design.
  13. Essay

    92 percent of Indonesian knowledge workers use AI. That is the floor, not the moat.

    Microsoft’s 2026 Work Trend Index made Indonesia the region’s adoption champion, then quietly reported the finding that undoes the boast.
  14. Essay

    Southeast Asia now holds 2,000 data centres. It captures almost none of the margin.

    The land, power and water are local; the models, the margin and the IP are almost entirely foreign, which makes the operator question simple: own a customer, not a GPU.
  15. Intel

    ByteDance's DeerFlow tops open-source rankings as agent harnesses become the layer

    DeerFlow, ByteDance's open-source superagent harness, pairs sandboxes, persistent memory, subagents and a message gateway into one model-agnostic runtime, and sits among the most-starred agent projects on GitHub.
  16. Intel

    GitHub Copilot moves to usage-based billing, metering agentic developer workflows

    GitHub replaced Copilot's flat premium requests with token-metered AI credits on June 1, and heavy agentic users report monthly bills climbing tenfold or more as repository-wide runs burn tokens fast.
  17. Intel

    Microsoft Agent Framework reaches GA, adding CodeAct and hosted agents

    The framework, consolidating AutoGen and Semantic Kernel into one open .NET and Python SDK, adds CodeAct, which has models emit a single Python program instead of sequential tool calls, cutting tokens 63.9% and latency 52.4% on tested workloads.
  18. Intel

    Open Design reaches 57,400 GitHub stars as local-first Claude Design alternative

    Eight weeks after its first commit, Open Design shipped v0.9.0 with 310 contributors, keeping design artifacts local via SQLite and auto-detecting 16 coding agents so teams avoid cloud lock-in.
  19. Intel

    Microsoft routes OpenAI models to China's tech giants through Singapore Azure

    ByteDance, Tencent, Ant Group and Meituan now reach OpenAI's GPT models through Azure regions in Singapore and Hong Kong, since Microsoft will not host them on mainland soil. ByteDance alone is on track to spend over US$1 billion a year.
  20. Intel

    OpenClaw passes 380,000 stars to become GitHub's most-starred project

    Peter Steinberger's local-first AI assistant crossed 380,000 stars and 79,500 forks, overtaking React in roughly sixty days. It runs on your own devices and answers through WhatsApp, Telegram, Slack and Discord rather than a hosted chat box.
  21. Intel

    Google donates A2A protocol to Linux Foundation for agent interoperability

    Google donated the A2A protocol to the Linux Foundation on June 23, 2025, establishing it as a vendor-neutral standard for AI agent communication across framework boundaries, with founding members including AWS, Cisco, Microsoft, Salesforce, SAP, and ServiceNow.
  22. Intel

    Google donates A2A agent-interop protocol to Linux Foundation under neutral governance

    Announced at Open Source Summit North America in June 2025, the A2A protocol became vendor-neutral in 2026 with founding members including AWS, Cisco, Microsoft, Salesforce, SAP, and ServiceNow. It enables agents to communicate across framework boundaries.
  23. Intel

    Microsoft AutoGen v0.4 rewrite triggers AG2 community fork over breaking changes

    Microsoft's AutoGen framework underwent a major v0.4 rewrite introducing substantial breaking changes that fragmented existing community implementations. Original v0.2 developers chose to maintain and develop the project under the new name AG2 rather than migrate to the rewritten codebase.
  24. Intel

    GitHub Copilot shifts to usage-based billing June 1, ending flat-rate developer AI

    All GitHub Copilot plans transition to token-based consumption pricing on June 1, 2026, replacing request-based limits with monthly AI Credits aligned to subscription tiers. Pro users get ten dollars in credits monthly; Pro+ users get thirty-nine dollars. Overage charges apply per-token beyond the allotment.
  25. Intel

    SaaS vendors layer AI consumption pricing atop subscriptions as 42% adopt hybrid models

    Microsoft 365 Copilot ranges from $27 to $43 per user per month with additional metered messaging; Atlassian bundles 25 Rovo AI credits per user then charges $0.30 per conversation overage; Salesforce Agentforce starts at $2 per conversation, reflecting industry-wide shift to tiered-plus-usage structures as AI features decouple from seat-based economics.
  26. Intel

    Singapore and Microsoft explore frontier AI model access for national safety and security testing

    Singapore is working with Microsoft to explore ways to access frontier artificial intelligence models for safety and security testing, according to The Straits Times report published June 13, 2026.
  27. Intel

    Google donates A2A protocol to Linux Foundation, forming A2A Project with AWS, Cisco, Microsoft, Salesforce, SAP, ServiceNow

    On June 23, 2025, Google officially donated the A2A protocol to the Linux Foundation at Open Source Summit North America in Denver, establishing it as a vendor-neutral standard for AI agent interoperability, with founding members including Amazon Web Services, Cisco, Microsoft, Salesforce, SAP, and ServiceNow.
  28. Intel

    Biome Rust-powered JavaScript toolchain gains enterprise CI adoption at 100x ESLint speed

    The zero-config linter and formatter written in Rust replaces ESLint and Prettier in a single binary, delivering sub-second lint times for large monorepos and cutting CI pipeline durations by orders of magnitude across production workloads.
  29. Intel

    Open Design reaches 57,400 GitHub stars in eight weeks as local-first Claude Design alternative

    Open Design hit version 0.9.0 on June 2, 2026 with 310 contributors, 6,500 forks, and 1,837 commits, offering self-hosting, model flexibility, and data-residency control for LLM-driven design artifact generation where Anthropic's Claude Design offers only cloud lock-in.
  30. Intel

    Microsoft ships seven in-house MAI models with reasoning, code, and Frontier Tuning

    Microsoft launched MAI-Thinking-1 reasoning model, MAI-Code-1-Flash for agentic coding at 5 billion active parameters, and four additional models for image, transcription, and speech on June 10. Frontier Tuning lets organizations train models on their own workflow traces, keeping institutional knowledge in-house while matching GPT-5.5 performance at 10x better cost efficiency.
  31. Intel

    Open Design hits 57,400 GitHub stars as local-first alternative to Claude Design

    Open Design reached v0.9.0 on June 2 with 310 contributors, offering model-agnostic design artifact generation, self-hosting, and data residency control. The tool auto-detects local coding agents, stores everything in SQLite, and supports Claude, OpenAI, Gemini, and Ollama endpoints with BYOK proxy and SSRF protection.
  32. Intel

    Google donates Agent-to-Agent protocol to Linux Foundation with AWS, Cisco, Microsoft, Salesforce backing

    Google officially donated the A2A protocol to the Linux Foundation on June 23, 2025, establishing it as a vendor-neutral standard for AI agent interoperability. Founding members include Amazon Web Services, Cisco, Microsoft, Salesforce, SAP, and ServiceNow, enabling agents to communicate across framework and organizational boundaries without single-provider lock-in.
  33. Intel

    GitHub Copilot transitions all plans to usage-based billing with AI Credits on June 1

    GitHub announced on April 27 that all Copilot plans will migrate to usage-based billing on June 1, replacing request-based pricing with token consumption. Pro and Pro+ monthly plans include $10 and $39 in monthly AI Credits respectively, with additional usage available for purchase. A preview bill experience launched in early May to give visibility into projected costs.
  34. Intel

    Microsoft ships seven in-house MAI models and ditches third-party distillation for clean training

    Microsoft released MAI-Thinking-1, MAI-Code-1-Flash, MAI-Image-2.5, MAI-Transcribe-1.5 and three on-device Aion models on June 2, all trained from scratch without distillation from OpenAI or Anthropic. MAI-Thinking-1 matches Sonnet 4.6 in blind evals while MAI-Code-1-Flash ships with 5 billion active parameters.
  35. Intel

    Microsoft releases Coreutils for Windows, NVIDIA RTX Spark Superchip, and OpenClaw native support

    At Build 2026, Microsoft shipped Coreutils for Windows (a Rust-based reimplementation of GNU Coreutils from the uutils open-source project), announced the Surface RTX Spark Dev Box with NVIDIA RTX Spark superchip delivering one petaflop of AI compute and 128 GB unified memory, and confirmed OpenClaw runs natively on Windows with MXC containment.
  36. Intel

    GitHub Copilot shifts all plans to usage-based billing and AI Credits on June 1

    All GitHub Copilot plans migrate to usage-based billing on June 1, 2026. Copilot Pro includes $10/month in AI Credits, Pro+ includes $39/month, and usage is calculated by token consumption (input, output, cached) at listed API rates. A preview bill experience launched in early May.
  37. Intel

    Microsoft 365 Business Standard with Copilot becomes permanent SKU at $23.50/user/month on July 1

    Promotional offers for Microsoft 365 Business Standard with Copilot and Business Premium with Copilot transition to permanent subscriptions on July 1, 2026, with updated list pricing of $23.50 and $32 per user per month respectively. The change removes the friction of selling Copilot as an add-on.
  38. Essay

    The 1.4 million deleted accounts that won’t fix the thing they were deleted for

    The June 4 disruption removed a million accounts and froze a few million dollars, and the durable casualty, the one no warrant reaches, is that everyone in a regional chat now half-assumes the other side is running a script.
  39. Intel

    Microsoft ships MAI-Thinking-1 reasoning model at 10× OpenAI cost efficiency for McKinsey workloads

    MAI-Thinking-1, a 35B active-parameter MoE with 256K context, hit 97% on AIME 25 and 53% on SWE Bench Pro while running 10× cheaper than GPT 5-5 on customer-tuned tasks. Available now via Microsoft Foundry for private preview.
  40. Intel

    China's OpenClaw hits 20 million monthly users and 250,000 GitHub stars in first month of mainstream adoption

    The open-source self-hosted AI agent reached 20 million MAU and became the fastest-growing open-source phenomenon in history, faster than Linux or React. Baidu, Tencent, and ByteDance all shipped localized versions in March; by June OpenClaw usage in China nearly doubled the US.
  41. Intel

    Microsoft ships MAI-Code-1-Flash to rival OpenAI at 10× the cost efficiency

    Microsoft unveiled MAI-Code-1-Flash, its inaugural coding model, at Build 2026 on June 2, claiming it outperforms GPT-5.5 with 10× better cost efficiency after refining for consulting-firm workloads like McKinsey's.
  42. Intel

    Open Design hits 57,400 GitHub stars in eight weeks as local-first alternative to Claude Design

    Open Design reached v0.9.0 on June 2, 2026, with 310 contributors and 57,400 stars. The project replicates Claude Design's artifact generation workflow but runs locally, supports any model via API proxy, and stores artifacts in SQLite—no cloud lock-in.
  43. Intel

    Microsoft AutoGen v0.4 rewrite triggers community fork; original developers launch AG2 as stable alternative

    Microsoft's AutoGen multi-agent framework shipped a v0.4 rewrite in June introducing breaking changes that fragmented existing implementations. Original v0.2 maintainers forked the project as AG2 to preserve stability and community continuity.
  44. Intel

    GitHub Copilot moves all plans to usage-based billing June 1, ending flat-rate subscriptions

    GitHub announced April 27 that Copilot will shift to token-based AI Credits on June 1, replacing per-seat pricing. Pro monthly plans include $10 or $39 in credits; Enterprise pooled credits align to seat count. Code completions remain unlimited; agentic workflows consume credits per token at varying model multipliers.
  45. Intel

    Microsoft embeds Copilot into Business Standard and Premium at $23.50 and $32/user from July 1, ending add-on friction

    Microsoft transitions Copilot from promotional add-on to permanent SKU embedded in Business Standard with Copilot ($23.50/user/month) and Business Premium with Copilot ($32/user/month) effective July 1. Pricing is durable; partners can position AI as core, not optional.
  46. Intel

    Microsoft Work IQ API launches with consumption-based Copilot Credits model, no separate subscription SKU

    Work IQ API reached general availability on June 16 with usage billed through Copilot Credits—a unified consumption currency across Microsoft Copilot Studio and AI services, replacing per-user or per-seat pricing.
  47. Intel

    Microsoft ships MAI-Code-1-Flash at Build, outperforms GPT-5.5 with 10x cost efficiency

    Announced June 2 at Build 2026, Microsoft's inference-optimized coding model runs in GitHub Copilot and Visual Studio Code, beating OpenAI's GPT-5.5 on McKinsey's benchmarks while delivering 10x better cost efficiency, according to Microsoft AI CEO Mustafa Suleyman.
  48. Intel

    GitHub Copilot shifts to usage-based billing June 1, ending flat-rate AI subscriptions

    Starting June 1, all GitHub Copilot plans transition from seat-based pricing to token consumption billed via AI Credits, with Pro users receiving ten dollars in monthly credits and Enterprise teams pooling usage across developers, ending the predictable flat-rate licensing era for AI coding tools.
  49. Intel

    GitHub Copilot moves all tiers to token-based credits June first

    Monthly flat subscriptions convert to AI Credit allotments priced at one cent per credit, consumed by input, output, and cached tokens. Pro users receive ten dollars in monthly credits; Enterprise pools usage across the organization.
  50. Intel

    Microsoft ships MAI-Code-1-Flash coding model, cutting OpenAI reliance by 10x cost efficiency

    At Build 2026, Microsoft launched MAI-Code-1-Flash and MAI-Thinking-1 reasoning models optimized for McKinsey's workflows, outperforming GPT 5-5 at one-tenth inference cost. Both ship in GitHub Copilot and Visual Studio Code, ending Azure's dependency on OpenAI for coding workloads.
  51. Intel

    Open Design reaches 57.4K GitHub stars in eight weeks as self-hosted Claude alternative

    Community project Open Design hit v0.9.0 on 2 June with 310 contributors, 1,837 commits, 6,500 forks. Offers local-first design-artifact generation, multi-provider API proxy (Anthropic, OpenAI, Azure, Gemini, Ollama), and Claude Design ZIP import. Runs via pnpm, deploys to Vercel, stores in SQLite.
  52. Intel

    GitHub Copilot switches to token-based billing, power users report 10x–50x cost increases

    As of 1 June, GitHub Copilot bills by AI Credits (tokenized consumption) instead of flat monthly fees. Base prices held, but developers running multi-turn agentic sessions face projected cost jumps of 10x–50x; some cancellations reported.
  53. Essay

    MTI raised the 2026 forecast. Manufacturing had its worst quarter since 2023.

    Singapore posted Q1 GDP up 4.6 percent year on year. Manufacturing fell 4.9 percent quarter on quarter, the steepest single-quarter drop since 2023. The AI hardware tailwind has rolled off.
  54. Essay

    What a ten-person Singapore startup will pay Microsoft to govern its AI agents.

    Microsoft moved Agent 365 to general availability on May 1, priced at fifteen US dollars per user per month. The governance layer is the small number. The compute it brackets is the large one.
  55. Intel

    Warp open-sources its agentic terminal under AGPL with OpenAI as founding sponsor

    The five-year-old Rust terminal moves to AGPL-3 with an agent-first contribution workflow and crossed 37,000 stars within days of release, hitting #2 on GitHub trending.
  56. Intel

    GitHub's Octoverse reveals friction-reduction tools dominating 2025 growth

    The fastest-growing languages and projects are no longer about shipping more code but reducing development friction as AI transforms how software gets built.
  57. Intel

    Trendshift launches engagement-based GitHub ranking beyond star counts

    New platform measures real repository engagement using Reddit and Hacker News discussion signals rather than vanity metrics.
  58. Essay

    Malaysia is becoming a jurisdiction. The question is which kind.

    Microsoft's $2.2 billion AI commitment to Malaysia is past the half-time mark. The certifications are the visible part of the deal. The data centres are the load-bearing part. Only one wins the strategic ledger.