Skip to content

AI Weekly · Public archive · September 15, 2026

Strong culture outperforms raw AI adoption every time.

It's culture. Here's the decision brief and a concrete experiment to run this week.

Sent to subscribers September 8, 2026. Public after the seven-day early-access window.

The AI noise is deafening, but the signal is clear. Your biggest productivity lever isn't a new model, it's the culture you build around it. This week, we cut through the turbulence to the one decision that matters.

The build decision

Prioritize culture over AI tooling

Before you budget for another AI seat, audit your team's psychological safety and communication clarity. A team with high trust and clear goals will outpace a dysfunctional team with the best AI tools. The trade-off is upfront work: you invest in people, processes, and clear documentation instead of chasing the latest model drop. Keep deterministic the core engineering rigor: code reviews, testing, and clear ownership. AI is a multiplier, not a foundation. It amplifies what's already there, good or bad. Fix the foundation first.

  1. 01

    AI creates a turbulent era

    Bill Gates warns that the rapid advancement of AI is creating a turbulent era requiring urgent global cooperation on safety, equity, and governance.

    Why it matters: Global governance must now keep pace with AI deployment.

    Source: gatesnotes.com

  2. 02

    Culture beats AI for productivity

    Strong engineering culture, trust, psychological safety, and clear communication, drives productivity more than AI tools.

    Why it matters: Trust and safety drive output more than any tool.

    Source: newsletter.eng-leadership.com

  3. 03

    Hugging Face's centralization risk

    Hugging Face's growing centralization in the open-source AI ecosystem raises concerns about control, sustainability, and bias, while new geopolitical alliances like the Five Eyes are formalizing AI governance, and space

    Why it matters: Open-source AI risks a single point of failure.

    Source: importai.substack.com

Field test

Run a one-question culture audit

  • 01Ask your team: 'On a scale of 1-10, how safe do you feel admitting a mistake?'
  • 02Share the anonymized aggregate score and discuss blockers in your next sync.
  • 03Pick one concrete blocker to fix this week, like clarifying a project goal or improving a review process.

Reader pulse

Your constraint can shape the next issue

What's one cultural fix that boosted your team's output more than any AI tool? Hit reply with one sentence, I read every response.

Seven days earlier

Get the next decision in your inbox.

One practical AI engineering brief every Tuesday. No feed to remember.

Subscribe