The Three Conversations Every Revenue Leader Needs to Have This Quarter About AI

You can't govern what you can't see.

Right now, your best people are using AI tools you didn't approve. They're uploading signed NDAs into Claude to summarize contract terms. They're feeding pipeline data into Perplexity to research competitors. They're pasting customer details into ChatGPT to draft follow-up emails.

They aren't being reckless. They're being resourceful. Because 70% of their day is admin drag, and AI is the only tool fast enough to keep them from drowning.

But here's the problem: you can't build governance around tools you don't know exist. You can't sanction what your team won't admit they're using. And you can't architect secure AI workflows when everyone's operating in the shadows.

Financial services faced this exact crisis 18 months ago. Banks tried to ban ChatGPT. Credit unions blocked Claude. It backfired spectacularly. Teams found workarounds. They used personal accounts. They accessed tools on mobile devices. Shadow AI got worse, not better.

So the smartest institutions changed tactics. They stopped trying to ban AI and started building what Dr. Angela Murphy calls "Amnesty and Orchestration" - a framework that makes it safe for teams to surface what they're already using, then helps leadership sanction the right tools with the right guardrails.

On the latest episode of The AI Hat Podcast, Angela - a PhD strategist known across FinTech as Payments Elsa - shared the exact conversation playbook financial institutions used to move from chaos to control in 30 days.

Here are the three conversations every Revenue Leader needs to have this quarter.

Conversation 1: "What Are You Already Using?" (The Shadow AI Amnesty Talk)

This is the hardest conversation - and the most critical.

You need to create amnesty. Make it safe for your team to admit they're using unsanctioned AI without fear of punishment or judgment.

Angela's script for financial services leaders works perfectly for revenue teams:

"I know many of you are experimenting with AI tools to get your work done faster. That's smart - you're being resourceful with the time you have. But we need to understand what's being used so we can make sure it's secure and compliant. This isn't about getting anyone in trouble. This is about making sure we're protecting our customers, our data, and your ability to keep using the tools that help you hit your numbers. So I need your help: what AI tools are you currently using, and what problems are they solving for you?"

The key is framing this as partnership, not interrogation.

When Angela ran this conversation with operations teams at a mid-sized credit union, she discovered something shocking. Nearly 60% of the back-office staff were uploading spreadsheets with customer transaction data into ChatGPT to search for specific records. They weren't trying to create a security breach. They were drowning in manual lookups that took 15-20 minutes per request, and AI cut that to 30 seconds.

Leadership's first instinct was panic. Angela's advice? Pause and listen.

"Don't react. Document. You're doing reconnaissance, not enforcement."

Ask follow-up questions:

• How often are you using this tool?
• What specific tasks does it help with?
• What data are you inputting?
• Why did you choose this tool over something else?

The answers will tell you two critical things: where your team's pain points are highest, and where you need sanctioned solutions immediately.

Conversation 2: "What Pain Points Are You Trying to Solve?" (The Workflow Audit)

Shadow AI doesn't emerge randomly. It appears where admin drag is crushing your team's capacity.

Angela asks every financial institution the same diagnostic question: "Where in your organization are people doing the most manual work?"

For revenue teams, that question translates to:

"What's taking the most time in your week that doesn't involve talking to customers or closing deals?"

Sit down with your top three performers. Not in a group setting - individual conversations where they feel safe being honest. Ask:

• What administrative tasks eat up the most hours?
• What parts of your job feel like unnecessary busy work?
• If you could automate one thing tomorrow, what would it be?
• What tools are you already using to try to solve these problems?

One VP of Sales Angela worked with discovered his team was spending an average of 8 hours per week updating CRM records, prepping for internal meetings, and pulling reports for leadership. That's nearly 20% of their capacity - gone to work that generates zero revenue.

When he asked what they'd already tried to fix it, three reps admitted they were using ChatGPT to draft meeting summaries and Perplexity to research accounts before discovery calls. They'd been doing it for six months. Leadership had no idea.

That conversation became the roadmap for what to sanction, what to train on, and what workflows needed redesigning.

Here's the key insight: Shadow AI is a diagnostic tool. It shows you where your processes are failing your people.

Don't treat it like a compliance violation. Treat it like free consulting on where to invest in automation.

Conversation 3: "Here's What We're Sanctioning and Why" (The Shadow AI Governance Rollout)

Once you know what your team is using and why, you can architect the solution.

Angela's framework for financial institutions works perfectly for revenue teams:

Step 1: Audit What's Already in Your Stack

Before you buy new tools, inventory what you already have. Most revenue teams are sitting on AI capabilities they don't know exist.

• Does your CRM have Einstein (Salesforce) or AI features you haven't activated?
• Do you have Microsoft 365 Copilot licenses gathering dust?
• Are there AI tools embedded in Gong, Clari, or your sales enablement platform?

One enterprise sales org Angela advised discovered they'd been paying for HubSpot's AI features for 18 months but never trained anyone on them. Their reps were using free ChatGPT instead - creating security risk and wasting budget.

The conversation: "We already have these AI tools available. Here's what they can do, and here's how to use them securely."

Step 2: Define What's Sanctioned (and What's Not)

Create a simple two-column list:

Sanctioned Tools:

• Microsoft Copilot (with organizational data controls)
• Salesforce Einstein (configured properly)
• [Internal GPT instance with privacy controls]

Not Sanctioned:

• Free ChatGPT accounts
• Claude.ai (public version)
• Uploading customer data to any non-approved tool

Don't just say "don't use these." Explain why.

The script: "We're not banning AI. We're making sure the AI you use protects our customers and our business. Free ChatGPT doesn't have the same data privacy controls as our enterprise tools. If you upload customer information there, it could be used to train their models or accessed in ways we can't control. That's not a risk we can take. But here's what you can use instead - and we'll train you on it this week."

Step 3: Commit to Ongoing Education

This is where most organizations fail. They announce a policy and assume everyone will figure it out.

Angela is adamant: "Education is not optional. It's the difference between adoption and resistance."

Financial institutions that successfully moved teams from Shadow AI to sanctioned systems did three things:

1. Brought in experts to train teams - Not IT explaining features, but practitioners showing real use cases
2. Created role-specific training - Sales reps need different prompts than RevOps analysts
3. Built feedback loops - Regular check-ins to ask "what's working, what's not, what else do you need?"

One bank brought Angela in to run prompt engineering workshops for their lending teams. The feedback was overwhelming: "This is the first time someone showed us how to actually use this stuff instead of just telling us it exists."

The conversation: "We're investing in training because we want you to be faster and better at your jobs - not because we're worried about compliance. We'll have workshops next week, and we'll bring in experts quarterly to keep you up to date. If you run into something these tools can't solve, tell us - that's how we'll know what else we need to add."

What Happens After These Conversations

Angela's seen this playbook work across dozens of financial institutions. The pattern is consistent:

Week 1: Initial resistance and fear. "Are we in trouble?" "Is this going to slow us down?"

Week 2: Relief and curiosity. "Wait, we can actually use these tools?" "Can you show me how to do [specific task]?"

Week 4: Adoption and advocacy. "I just saved 3 hours on a proposal." "Can we get access to the advanced features?"

The teams that succeed are the ones where leadership frames AI governance as enablement, not enforcement.

You're not taking tools away. You're giving them better ones - with training, support, and security they can defend.

Your Monday Morning Shadow AI Action Plan

If you're a VP of Sales, CRO, or RevOps Director who suspects Shadow AI is happening on your team, here's how to start:

This Week:

1. Schedule 1:1s with your top three performers. Use the "What are you already using?" script.
2. Document what you learn without reacting. You're gathering intelligence, not issuing warnings.

Next Week:

3. Audit your existing tech stack. What AI is already available that your team doesn't know about?
4. Draft your "Sanctioned vs. Not Sanctioned" list with clear explanations of why.

Week 3:

5. Announce your AI governance framework in a team meeting. Lead with amnesty, not consequences.
6. Bring in an expert (internal or external) to run training on your sanctioned tools.

Week 4:

7. Launch feedback loops. Create a channel where people can ask questions, report issues, and request additional capabilities.

Angela's advice: "Don't wait for perfect. Start with good enough and iterate. The cost of doing nothing is exponentially higher than the cost of imperfect governance."

The Window Is Closing

Financial services didn't have a choice. Regulators forced the governance conversation. The GENESIS Executive Order, the EU AI Act, and increasing scrutiny around data privacy made amnesty and orchestration a survival requirement.

You still have a 12-18 month window before those same pressures cascade to every B2B industry. You can architect governance now as a competitive advantage - or scramble to build it later under regulatory pressure.

The three conversations aren't compliance theater. They're leadership in the Agentic Era.

Your team is already using AI. The question is whether you're architecting the solution or pretending the problem doesn't exist.

Resources

Listen to the full episode with Dr. Angela Murphy:
Why Your Sales Team's Shadow AI Is a Lawsuit Waiting to Happen - A Fintech CRO's Governance Playbook

Download the Executive Guide to Shadow AI:
Get the framework for turning security risk into competitive advantage
theaihat.com/shadow-ai

Connect with Angela Murphy (Payments Elsa):
LinkedIn: linkedin.com/in/paymentselsa

Need help running these conversations with your team?
I deliver custom Revenue Architecture workshops for VPs of Sales, CROs, and RevOps leaders who need practical AI governance frameworks - not abstract theory. Let's chat.