Search

The Bigger Picture: AI Is Coming Faster Than Policies Can Keep Up

Michael Lee, MBA
May 2
3 min read

Organizations are scrambling to write AI use policies. But here’s what many miss:

Policy without culture is paper. Policy without literacy is risk.

Organizations are racing to adopt AI, but many are doing so without fully preparing their people.

According to the IBM Global AI Adoption Index 2023 (PDF):

42% of enterprise-scale organizations (with over 1,000 employees) report having actively deployed AI.
An additional 40% are exploring or experimenting with AI tools in pilot programs.
However, 34% cite limited AI skills or expertise as a top barrier to adoption.

This highlights a critical gap: companies are investing in AI systems but often underinvesting in human readiness. The lack of structured employee education, particularly around ethical use and prompt design, leaves room for misuse — not out of malice, but out of ignorance.

💼 The Leaders Who Get It — and Those Who Don’t

Two paths are emerging:

❌ The Restrictors

Some firms (especially in finance, healthcare, and legal) have banned generative AI tools outright. JPMorgan Chase, for instance, temporarily blocked ChatGPT citing data security risks.

But bans often drive usage underground, increasing exposure, not reducing it.

✅ The Enablers

Others — like PwC, Morgan Stanley, and Accenture — have embraced AI with controls:

They give employees access to enterprise-grade AI tools.
They run workshops on ethical AI use.
They encourage open sharing of use cases and wins.

Accenture even created an internal “AI Assistance Guild” to share safe practices and case studies across business units. That’s smart governance.

🔧 5 Moves to Manage AI the Right Way

Here’s what forward-looking organizations are doing (or should be):

1. Make It Safe to Talk About AI

Encourage teams to share:

“How I used AI this week”
“A mistake I caught”
“A result I was proud of”

Normalize the conversation — make learning visible.

2. Establish Guardrails, Not Walls

Instead of bans, give access to tools like:

Microsoft Copilot for Microsoft 365, which integrates into your existing Office environment while maintaining data privacy, respecting user permissions, and enabling audit tracking via Microsoft Purview.
OpenAI Enterprise (ChatGPT Enterprise), which ensures your data is not used for training, provides SOC 2 Type 2 compliance, and offers admin dashboards for usage and access management.
Google Gemini for Workspace, which processes data securely within the Workspace environment, adheres to Google's privacy commitments, and offers administrator-level tracking for responsible use.

All three tools support data protection, usage monitoring, and content governance — giving teams the power of AI without the risks of shadow IT or uncontrolled public tool use.

3. Upskill Everyone with AI Literacy

A simple framework like PEACE can guide better AI use:

Purpose – What are you trying to achieve?
Example – Provide a good prompt or format.
Audience – Who is the output for?
Context – What information matters?
Expectations – Tone, depth, style?

Add to this a discussion of Responsible AI Principles:

Fairness
Transparency
Accountability
Privacy

4. Reward Transparency

Treat AI use like you would spreadsheet macros or slide templates — if it helped, cite it. Encourage honesty.

5. Train Leaders First

Many execs haven’t used AI firsthand. Train them first, so they can speak credibly and lead cultural change.

🧭 Where Are We Headed?

The future isn’t just AI-powered — it’s AI-normalized.

Like email, PowerPoint, or Excel, generative AI will soon become a base-level competency. But for now, it sits in a confusing place: wildly powerful, poorly regulated, unevenly understood.

That’s why the real challenge isn’t AI — it’s culture.