Like it or not, AI is already in your workplace. The tools are being used — quietly — by individuals across teams, even if no one’s officially said a word.But in many organisations, AI has become the elephant in the boardroom. There’s no clear direction. No open conversation. Just solo experiments, vague encouragement, and silent uncertainty.Related Article Block PlaceholderArticle ID: 299148Two in three Aussie office workers using AI tools without telling the bossJennifer Dudley-NicholsonAnd when we don’t talk about AI, we let confusion, inconsistency, and risky habits take root. Culture forms by default, not by design.It’s time to break the silence. Because awkward or not, this is where real leadership begins.The risk of keeping AI in the shadowsJust because no one’s talking about AI doesn’t mean it’s not being used. And when it’s happening quietly, without guidance or guardrails, things start to unravel.You get patchy quality. Conflicting results. Silent stress from team members who don’t know if they’re ‘doing it right’ — or if they’re evenallowedto try.This kind of “don’t ask, don’t tell” culture doesn’t keep things safe. It just keeps things messy. People either overstep, underperform, or avoid AI altogether — not because they’re lazy, but because no one’s made the expectations clear.In my bookAI-Human Fusion, I talk about the HABITS framework, where theHstands forhumans first. Before diving into tools or workflows, we need to create a space where curiosity is encouraged and questions aren’t a sign of weakness because confident AI use starts with psychological safety.Avoiding the conversation won’t minimise the risks. It just guarantees you won’t see them coming.Related Article Block PlaceholderArticle ID: 311872Lawmakers call for mandatory AI guardrails to prevent private sector RobodebtDavid AdamsAI culture is forming — with or without youWant to bring AI out of the shadows without overwhelming your team? Start with honesty and curiosity.At your next team meeting, try something like:We know AI’s already part of how we work – even if we haven’t talked about it much yet. Let’s figure out how to use it well, together.That one sentence can shift the tone. It signals that AI isn’t off-limits, it’s okay not to know everything, and there’s room to explore.From there, invite input: run a quick survey, start a shared doc, or host a casual chat. Ask simple questions like:What tools are you already using?What’s been helpful – or confusing?Where could AI actually save us time?Then listen. Your team doesn’t need perfection. They need permission to learn, try, and talk about it openly.Because if you don’t guide the culture, the tools will do it for you.Turn AI curiosity into confident actionOnce the conversation’s out in the open, it’s time to add some structure. Not rigid rules, but gentle scaffolding that helps your team learn, grow, and stay aligned as they experiment.Related Article Block PlaceholderArticle ID: 315394Shopify CEO’s bold AI memo a lesson for other executivesDavid BrudenellHere are three simple (but powerful) ways to build capability without overwhelming anyone:AI buddy programsPair up your confident explorers with your cautious observers. Let them share tips, try prompts together, swap lessons, and reflect on wins. Learning in pairs lowers the stakes and makes the journey feel less lonely.Monthly AI town hallsSet up a regular meeting to bring the whole team together. Share what’s working. Workshop new ideas. Celebrate creative use cases. And normalise the idea that AI isn’t just for techies — it’s for everyone.AI hackathonsThese aren’t just for coders. Run a short, high-energy challenge where teamstrial AI on real business problems— like speeding up reporting, brainstorming marketing angles, or refreshing the onboarding process. Keep it hands-on, low-risk, and fun.Related Article Block PlaceholderArticle ID: 314291Back to basics: Culture Amp doubles down on hackathons to upskill engineersMelissa IariaThese programs are about building a culture of shared learning and experimentation. They help connect the dots between curiosity and real-world value, so your team doesn’t just tinker with AI, but actually uses it to think better, move faster, and solve meaningful problems.Set clear boundaries before innovation goes rogueExploration is great. But without a few guardrails in place, things can veer off course fast. That’s why, once the experimenting begins, it’s time to set some shared expectations. Not in a ‘red tape’ way, but in a way that protects your team and gives everyone clarity.Start with questions like:What tasks is AI allowed to support and which are off-limits?How should we disclose AI use in client-facing work?Who’s responsible for checking and approving AI-generated output?These aren’t rules to box people in. They’re shared agreements that keep everyone moving in the same direction.As I highlight in the BI and TS pillars of my HABITS framework (that’s Business Integration and Tactical Strategy), innovation is only valuable when it’s aligned, intentional, and sustainable. Otherwise, you end up with a pile of half-baked outputs and no one is quite sure who owns what.Boundaries don’t kill creativity. They make space for it.Conversations lead to capabilityAI isn’t going anywhere. But how it’s used — and how it feels to use it — is still completely within your control.By breaking the silence now, you lay the foundations for clarity, confidence, and collaboration. Because that awkward AI conversation you’ve been putting off is actually your biggest opportunity for long-term success.Start small. Ask the right questions. Create space for transparency.Your future workflows — and your people — will thank you for it.Never miss a story: sign up toSmartCompany’sfree daily newsletterand find our best stories onLinkedIn.