It Just Dawned On Me Series No. 4

Empathy Caught It Early -- And Saved Us Later

It just dawned on me that empathy isn’t soft. It’s smart. It’s strategic.

In AI-powered workplaces, like law firms, it might be the only thing standing between a small workaround and a massive risk event.

Here’s how one quiet pause in a training session changed everything.

Empathy Caught It Early

I was advising a midsize professional services firm exploring ChatGPT Teams to help employees draft client reports more efficiently. A smart move, until a small moment shifted everything.

During a training session, I noticed a junior employee looking uneasy. So I paused and asked:

“What’s been working for you so far?”

That single moment of empathy, just giving space, led to a big reveal:

Team members had already been using their own, personal paid ChatGPT accounts to speed things up.

They copied and pasted sensitive client data -- names, dates, financials -- into unsecured tools.

Situational Empathy  

Empathy in that moment meant no blame, no shame, only foresight.

It caught the risk early, earned trust, and built stronger guardrails before something broke.

Here’s the Problem

Those accounts weren’t part of the company’s security system:

🚫 No audit trail.

🚫 No control.

🚫 No governance.

The secure ChatGPT Teams platform hadn’t even gone live, but the risks were live.

Imagine one employee clicking a phishing link or having malware on their laptop.

The breach wouldn’t come from hackers. It would come from habit.

Not malicious, just human behavior. A well-intentioned workaround under pressure.

What We Did Next

Empathy did not lead to blame or shame; it led to better governance.

We paused the rollout and built the right foundation:

✅ Clear rules on what can and can’t go into AI tools.

✅ Pre-approved prompt templates.

✅ Systems that track who accesses what, and when.

✅ Secure storage for AI-related documents (not in inboxes or desktops).

Here’s the Truth

If you can’t trace it, you can’t contain it.

Lesson Learned

When something goes wrong -- and it will -- you need to show your work:

✅Who touched the data?

✅What tool was used?

✅When did it happen?

That’s your safety net.

Even Paid GPT Accounts & Settings Pose Risks

1. Device-Level Risk

If your team member is using a personal device (laptop, tablet, or phone):

• You cannot verify whether that device is encrypted.

• You do not control whether it is infected with malware.

• There is no audit trail for data input (what, when, or by whom).

🚨 Risk: Even if OpenAI doesn’t store the data, a keylogger or screen capture malware could compromise it locally.

2. No Enterprise Logging or Oversight

Paid accounts are individual-facing, not enterprise-managed.

• No visibility into prompts or outputs.

• No version control or documentation.

• Can’t fulfill discovery, compliance, or audit requirements.

🚨 Risk: In legal or regulatory contexts, this becomes a "rogue tool use" issue, not aligned with company controls.

3. Data Leakage via Copy-Paste Habits

If an employee:

• Pastes sensitive data into ChatGPT.

• Copies the AI response.

• Saves it into an unsecured location (local drive, desktop, email).

🚨 Risk: You now have data fragments outside any protected system, vulnerable to mishandling or breach.

4. Misunderstanding of the “Do Not Train or Share” Setting

Even if you turn on the setting that says “Don’t use my content to train the model":

• It doesn't mean the information is fully protected.

• It doesn’t follow your company’s privacy rules.

• It doesn’t give you the same legal protection as using ChatGPT Teams or Enterprise.

🚨 Risk: That setting only stops the AI from learning from your input. It doesn’t protect the data you put in, or how it’s handled afterward.

Bottom Line:

Paid is not secure. Personal is not controlled.

Even well-meaning employees using ChatGPT Plus with privacy settings still introduce risk if:

✅They’re handling sensitive or regulated data.

✅They’re working outside company-governed platforms.

✅The organization hasn’t issued guidance, governance, or secure alternatives.

The Impact of Empathy

Empathy sharpens our response.

It brings people in without fear or shame.

It’s not just kindness, it is diligence, trust, and smart leadership in action.

Empathy influences company culture.  

Empathy isn’t a nice-to-have. It’s a signal of serious leadership in the age of AI.

It’s how resilient cultures are built—and how risks get caught before they become headlines.

"Empathy isn’t soft. It’s strong. It brings people in, builds trust, and protects what matters before it breaks."

What’s one AI risk or blind spot you’ve caught early, or wish you had?

Let’s discover them together.

#AILeadership #Empathy #Governance #AIRisk #PrivacyRisk #ItJustDawnedOnMe #TheCyberDawn