Every post is adapted from our newsletter and tuned for clarity, so you can scan, share, and act on the ideas that move Canadian workplaces forward.
Weekly newsletter
Want the TL;DR in your inbox first? Subscribe to the early week drop.
Subscribe to the weekly blogTwo months of building at Suprvisr: reusable Skills that run whole tasks on their own, a Skill Creator that writes them with you, skill imports from GitHub, and spreadsheets you can question and build in chat. Plus how Execulink Telecom's staff use Cumulus every day.
OpenAI's models escape a test sandbox and breach Hugging Face, ChatGPT Health opens to all US users amid a lawsuit, the White House accuses Moonshot of distilling Anthropic's Fable, and SAP buys Prior Labs.
China launches a 29-country AI governance body as US labs publish rival rulebooks, Apple sues OpenAI over trade secrets, Moonshot gives away Kimi K3, and Anthropic backs a $1.5B bet on implementation.
GPT-5.6 clears its government review and GPT-Live talks like a person, an AI agent runs most of a ransomware attack, Anthropic out-earns OpenAI, and Google trains on your uploads unless you opt out.
Fable 5 and Mythos 5 return after a 19-day blackout with a new severity framework, Claude Sonnet 5 launches at aggressive pricing, Cloudflare makes AI crawlers pay, and Meituan open-sources LongCat-2.0.
GPT-5.6 launches behind a government gate, Gemini 3.5 Pro misses June, Google DeepMind loses four researchers in 48 hours, China open-sources GLM-5.2, and $2.7 trillion leaves the AI trade.
Thirteen words in a Reddit comment can hijack AI research tools, only 16% of Americans expect AI to help society, SpaceX buys Cursor for $60B, and a one-click Copilot flaw leaked 2FA codes.
A government order pulls Anthropic's Fable 5 offline three days after launch, SpaceX makes Musk the first trillionaire, Siri lets you pick a different model, and OpenAI buys the plumbing under its agents.
Anthropic says AI has started building itself and asks for a coordinated brake, Canada launches its AI for All strategy, Microsoft ships its own models, and four rival labs agree on DNA screening.
Anthropic passes OpenAI at a $965B valuation, Microsoft ships agents that can use any app, Apple's Siri borrows Google's Gemini, and Beijing starts keeping its AI talent at home.
Google turns Search into an agent, OpenAI files for IPO after winning the Musk lawsuit, Trump pulls his own AI executive order, and Andrej Karpathy joins Anthropic.
Anthropic rents Elon's data center, Claude agents learn to dream, GPT-5.5 trades emojis for accuracy, the goblin debugging confession, and TELUS gets a public lesson in disclosure.
Microsoft and OpenAI rewrite their exclusivity deal, a 9-second AI agent disaster wipes a production database, Big Tech bets a trillion dollars on AI infrastructure, and the Mercor breach makes voice authentication officially indefensible.
Google Cloud Next 2026 launches autonomous agent platform, OpenAI ships three products in five days, DeepSeek V4 raises the open-source bar, Gemini Embedding 2 goes GA, Canada's PMO replaces staff with AI, and BDC's $500M SME push.
Claude Opus 4.7 launches and burns credits fast, Allbirds pivots to GPUs, Canada opens $890M sovereign AI compute applications, Stanford's 2026 AI Index shows mainstream adoption, Bank of Canada meets on Anthropic's Mythos, and leaders start building their own agents.
Anthropic's Claude Mythos Preview restricted to 40 vetted organizations, the Fed and Treasury brief bank CEOs on AI cybersecurity risks, Google DeepMind's TurboQuant cuts AI memory costs 6x, Meta goes closed-source with Muse Spark, and OpenAI publishes its superintelligence policy wishlist.
Anthropic accidentally published 512,000 lines of Claude Code source to npm, OpenAI closed an $852B valuation round, Google released Gemma 4 as truly open, Alibaba shipped a 1-million-token model for free, and Gartner says copilots are dead by 2028.
OpenAI shuts down Sora and loses Disney, Anthropic hits $19B revenue, NVIDIA Vera Rubin and NemoClaw, Anthropic rate limits, White House copyright policy, Google Gemini 3.1 Flash, Deloitte AI hallucinations, and Ontario's humanoid robotics corridor.
Anthropic blacklisted by the Pentagon, OpenAI takes the contract, 295% ChatGPT uninstall surge, Nano Banana 2 launch, GPT-5.4, AI agent oversight gaps, and building safe AI culture in Canadian enterprises.
The Moltbook security scandal, SpaceX acquires xAI for $1.25 trillion, AI internal mumbling research, China's Kimi K2.5, and Canada's agentic AI metrics — weekly AI insights for Canadian business leaders.
Apple partners with Google Gemini for Siri, Yann LeCun exits Meta, Ilya Sutskever's $117B paper wealth, Canada's AI talent exodus, Pentagon adopts Grok, and Boston Dynamics Atlas goes to work.
CES 2026 hardware recap, xAI's $20B scandal, Nvidia acquires Groq, California vs federal AI regulation, and ChatGPT Health — Canadian AI insights for enterprise leaders.
Holiday edition: playground toast, model race, IBM-Confluent, TIME's AI architects, MCP to Linux, DIY AI, and holiday stats.
Playground access, brand caution, the GPT-5.2 debate, Disney's Sora bet, legal blowback, and open-source momentum.
OpenAI hits code red, Meta wires news into your chats, sovereign AI goes big in Australia, and RAM prices climb.
Mental health bots pull back, agents start paying bills, incentives get messy, and Ottawa ships a federal AI register.
Google ships Gemini 3 and an AI-first IDE, Cloudflare reminds everyone who runs the internet, Congress wakes up to GenAI risk, and spies automate with Claude.
Nearly every enterprise is piloting AI, but only the governed ones see ROI. Here's how leaders can close the gap.
Canada doubles down on sovereign AI, Europe blinks, and Big Tech signs a $38B handshake. Here's what it means for leaders.
Atlas launches with red flags, Ottawa wraps its AI sprint, and Canadian leaders double down on data sovereignty.
Comfortably numb: half of Canadians rely on AI at work, even as trust and governance lag behind.
Canada's slow start in AI might be the smartest move yet - a governance-first approach that could define the next phase of innovation.
The cost of bad AI governance is rising fast - from data exposure to brand damage. Here's how leaders can stay ahead.