Two Major Studies Just Proved AI Creates Jobs. Here's the Catch SMBs Are Missing.

Two studies dropped in the same week and told completely different stories about AI and jobs. Both are right. And if you run a small business, the gap between them is where your risk lives.

The European Central Bank surveyed nearly 5,000 firms and found that companies investing in AI are 4% more likely to add staff than those that aren’t. Two-thirds of firms reported employees already using AI. The headline: AI creates jobs.

Three days later, the International Labour Organization published a research brief on generative AI and gender that paints a harder picture. 29% of female-dominated occupations face GenAI exposure compared to just 16% of male-dominated ones. When you zoom into high-automation risk categories, it’s 16% vs. 3%.

Same technology. Same week. Two radically different outcomes depending on who you are and what you do.

---

Quick Verdict

AI is a net job creator right now. But the benefits aren’t distributed evenly, and the ECB warns the positive employment effect may be temporary. SMBs that ignore the gender disparity in AI exposure risk losing their best administrative and operations talent to automation anxiety. The window to get ahead of both trends is this quarter, not next year.

---

What the ECB Actually Found

The ECB’s survey is the largest employer-side study on AI and jobs we’ve seen in 2026. Here’s what matters for business owners.

Finding	Number	What It Means for SMBs
Firms with employees using AI	66%	AI adoption is mainstream, not experimental
AI-investing firms more likely to hire	+4%	Investing in AI correlates with growth, not cuts
Large firms (250+) using AI	~90%	Enterprise adoption is near-universal
Small firms (<10) using AI	~60%	SMBs are still 30 points behind
Future hiring expectations	Positive	AI-investing firms plan to keep hiring

The 30-point adoption gap between large and small firms is the number that should worry you. Almost every company with 250+ employees already uses AI. If you’re running a 10-person shop and haven’t deployed AI in daily operations, you’re competing against organizations that have a structural efficiency advantage you can’t see on a P&L yet.

I’ve written about this adoption gap before. The AI adoption drives small business growth data backs this up: SMBs that invested in AI tools reported measurable revenue gains. The ECB data confirms it at European scale.

But here’s the caveat the headlines skip. The ECB blog explicitly states that the positive employment effect may be temporary. Their reasoning: AI hasn’t yet significantly transformed production processes. Companies are adding staff to manage AI implementation. Once AI fully replaces the workflows those staff manage, the hiring picture could reverse.

That’s not fear-mongering. That’s the ECB telling you to use this hiring window strategically, not assume it lasts forever.

---

The Gender Gap Nobody in Tech Is Talking About

The ILO brief landed on March 5 and got a fraction of the attention the ECB study received. That’s a mistake, because the gender data has direct implications for every business with administrative, operations, or customer service teams.

Here’s the core finding: women face nearly twice the AI automation exposure that men do. Not because of capability differences, but because of where women work.

Why women are disproportionately exposed:

• Women are concentrated in clerical, administrative, and business support roles (secretaries, receptionists, payroll clerks, accounting assistants) where tasks are routine and automatable
• Women remain underrepresented in AI-related and STEM occupations that build and manage the systems doing the automating
• AI systems often reproduce gender biases embedded in training data

At the country level, women face higher GenAI exposure than men in 88% of countries analyzed. In economies like Switzerland, the UK, and the Philippines, more than 40% of women’s employment is exposed to generative AI.

If your office manager, bookkeeper, or customer service lead is a woman handling routine administrative tasks, she’s in the highest-exposure category. And she probably knows it, even if you haven’t talked about it.

---

Why This Matters to Your 15-Person Company

I keep hearing SMB owners say some version of “this is an enterprise problem.” It isn’t. Here’s why both studies hit small businesses harder.

The ECB adoption gap is an SMB problem. 90% of large firms use AI. 60% of small firms do. That 30-point gap means your larger competitors are already capturing the efficiency gains you’re still evaluating. Every quarter you wait, the gap compounds. I broke down this compounding effect in the AI skills crisis analysis, and the math hasn’t changed.

The gender gap is an SMB retention problem. Small businesses rely heavily on administrative staff who handle multiple functions. Your office manager who does bookkeeping, scheduling, vendor management, and HR paperwork? She’s doing exactly the work GenAI targets first. If you don’t proactively retrain and reposition her role, one of two things happens: she leaves for a company that’s invested in her growth, or you automate her out and lose institutional knowledge you can’t replace.

Neither outcome is good. Both are preventable.

The temporary hiring window is an SMB strategy problem. The ECB says AI firms are hiring now, but warns it might not last. Large companies can absorb that shift. They have workforce planning teams and retraining budgets. You have a spreadsheet and 90 days of cash runway. If you’re hiring to manage AI implementation, plan now for what those roles become once the implementation is done.

---

The Playbook: How SMBs Should Respond to Both Studies

I’ve spent the last 18 months helping small businesses navigate exactly this kind of contradictory data. Here’s the framework that works.

1. Audit Your AI Adoption Against the ECB Benchmarks

If fewer than 66% of your employees use AI tools in daily work, you’re below the European average. For a 15-person company, that means at least 10 people should have AI integrated into their workflows.

Not “have access to ChatGPT.” Integrated. Using AI tools to complete actual work tasks faster and better than they could without AI.

Run this audit this week. It takes 30 minutes. Ask every employee two questions: “Do you use AI tools for work?” and “Which specific tasks?” The gap between “access” and “use” will surprise you.

2. Map Your Gender-Exposure Risk

Look at your team. Identify every role that involves routine clerical, administrative, or data-processing tasks. Cross-reference with who fills those roles. If your highest-exposure positions are held by women (and statistically, they will be), you have a retention and retraining priority.

This isn’t about corporate diversity programs. This is about business continuity. When your most automatable roles are concentrated in a subset of your team, you need a plan that transitions those people into higher-value work before the automation arrives, or before they leave because they see it coming.

The positioning for AI job market changes guide covers the individual career side. As a business owner, your job is to create those transition paths internally.

3. Invest During the Hiring Window (But Plan for What Comes After)

The ECB data says AI-investing firms are hiring. Good. Use this window to bring in people who can help you deploy AI across your operations. But structure those roles with built-in evolution.

Don’t hire an “AI implementation coordinator” as a permanent role. Hire someone to implement AI over 6-12 months, then transition them into the operations or strategy work that AI frees up. The temporary nature of the hiring bump means every new role needs a post-implementation purpose.

4. Address the Anxiety Before It Becomes Attrition

The ILO data is public. Your employees can read it. The women on your team in administrative roles now know that 29% of female-dominated occupations are exposed to GenAI. If you don’t address this directly, they’ll fill the silence with worst-case assumptions.

I’ve seen this play out with clients. The double-edged reality of AI at work piece covers how productivity gains and job anxiety coexist in the same workforce. Ignoring the anxiety doesn’t make it go away. It makes your best people update their resumes.

Have the conversation. Be specific about how you plan to use AI to enhance roles, not eliminate them. Back it up with a retraining commitment and a timeline.

---

What the ECB’s “Temporary” Warning Really Means

The ECB didn’t bury this point. They put it in the blog post because they know the headline “AI creates jobs” will get misread as “AI will always create jobs.”

Here’s what they actually said: most surveys that show pessimistic employment effects cover longer timeframes. The ECB survey captures the current moment, which happens to be the implementation phase. Companies are hiring to build and deploy AI. Once those systems run autonomously, the hiring rationale changes.

Think of it like construction. Building a factory creates construction jobs. Running the factory doesn’t employ construction workers. We’re in the “building” phase of AI deployment. The enterprise AI ROI reckoning covers how companies are already starting to measure whether their AI investments justify ongoing headcount.

For SMBs, this means the next 12-18 months are a window. Hire smart. Deploy fast. Build the systems that will generate returns long after the implementation phase ends. If you wait until the hiring window closes, you’ll be competing for talent in a market that’s already shifted.

---

Three Numbers to Track This Quarter

Forget vanity metrics. Based on both studies, here are the three numbers every SMB should track through Q2 2026:

1. AI tool usage rate across your team — Target: 66%+ of employees using AI for specific daily tasks (ECB benchmark)
2. Automation exposure by role and gender — Map which roles face the highest GenAI exposure and who fills them (ILO framework)
3. Productivity per employee before and after AI deployment — This is how you prove the ECB’s hiring correlation holds for your business

If number one is below 66%, you’re behind the curve. If number two shows concentrated risk in a demographic subset of your team, you have a retention problem forming. If number three isn’t moving within 90 days of deployment, your AI implementation needs rework.

---

The Bottom Line

The ECB proved AI creates jobs today. The ILO proved it doesn’t create them equally. And the ECB’s own researchers warned the positive trend might not last.

For SMBs, all three findings point to the same action: deploy AI now, retrain your most-exposed employees (especially women in administrative roles), and build roles that have purpose beyond the implementation phase.

The businesses that read both studies and act on the full picture will be positioned for whatever comes next. The businesses that only read the “AI creates jobs” headline will be blindsided when the temporary hiring window closes.

Your next step: Run the 30-minute team audit from the playbook above. Two questions per employee. Map the answers against both the ECB adoption benchmark and the ILO exposure categories. That single exercise gives you more actionable intelligence than any amount of headline reading.