What Small Businesses Can Learn from Big Companies' AI Experiments

Introduction

Artificial intelligence is no longer a futuristic buzzword; it’s a daily operating tool for enterprises of every size. While the headlines often focus on the multi‑billion‑dollar AI projects of corporations like Meta, Oracle, and OpenAI, the real value for small and mid‑sized businesses (SMBs) lies in the lessons those giants leave behind. Gene Marks points out that the massive investments—both successful and squandered—by big brands are giving SMBs a free playbook on what works, what doesn’t, and where to spend (or not spend) their limited AI budgets.

• --

What to Do: Proven Strategies from the Big Players

1. Use AI as an analytical assistant, not a replacement

• Large firms have struggled with the reliability of “agentic” AI that claims to handle entire functions. SMBs are wiser to keep core accounting, marketing, and customer‑service teams while letting AI surface insights, recommend strategies, and automate routine data‑processing.

2. Beta‑test new tools through the lens of a pilot

• Treat every new AI service as a pilot project. Track key performance indicators (KPIs) before scaling. This mirrors how corporations roll out agentic tools internally before a full‑scale launch.

3. Diversify AI services

• Figure 4 from the JPMorgan study shows SMBs moving from a single AI service (89 % in 2019) to multiple concurrent services by 2025. A diversified stack reduces reliance on any one vendor and spreads risk.

4. Focus on revenue‑linked adoption

• Adoption rates climb with revenue size. By December 2025, firms with > $250 k revenue adopt AI at 22.5 % versus 15.1 % for smaller peers. Target AI projects that directly impact top‑line growth to justify the expense.

5. Leverage AI for predictive analytics and smarter ERP

• Industries such as steelmaking demonstrate AI’s power in robotics, predictive maintenance, and safer workplaces. SMBs can apply similar analytics to inventory, cash‑flow forecasting, and sales pipeline management.
• --

What Not to Do: Common Pitfalls to Avoid

| Pitfall | Why It Fails | SMB Mitigation |
|---|---|---|
| Chasing trends without a clear use‑case | Leads to wasted spend and data sprawl. | Start with a business problem, then match the AI tool that solves it. |
| Dumping sensitive data into unsecured tools | Risks regulatory breaches and brand damage. | Use vetted, GDPR‑compliant platforms; encrypt data at rest and in transit. |
| Assuming AI will replace staff | Studies show AI is more a productivity lever than a layoff driver; firms often use AI as a distraction from weak demand. | Communicate AI’s role as augmentation; retrain staff to work alongside the technology. |
| Relying on a single vendor | Creates vendor lock‑in and single‑point failure. | Adopt a multi‑vendor approach as shown by the rise in multiple‑service adoption. |
| Neglecting ongoing monitoring | Agentic AI can drift, producing inaccurate outputs. | Implement continuous validation loops and human‑in‑the‑loop oversight. |

• --

Adoption Trends: Numbers That Matter

| Year | % of Small Firms (< $250 k) Using AI | % of Large Firms (> $250 k) Using AI |
|------|--------------------------------------|--------------------------------------|
| 2019 | 9 % | 14 % |
| 2021 | 12 % | 18 % |
| 2023 | 14 % | 20 % |
| 2025 (proj.) | 15.1 % | 22.5 % |

Source: JPMorgan Institute, “Understanding the Use of AI Among Small Businesses”【https://www.jpmorganchase.com/institute/all-topics/business-growth-and-entrepreneurship/understanding-ai-use-by-small-businesses】

The chart illustrates a steady, though modest, rise in AI uptake among SMBs. The gap between large and small firms persists, underscoring the importance of targeted, ROI‑driven AI projects for smaller players.

• --

Avoiding the “Debt Bomb” Narrative

Some experts warn of a looming “AI debt bomb” as corporations over‑invest in data‑center capacity (Meta, Oracle, etc.). For SMBs, the risk is far lower because:

• Capital expenditures are modest.


• Most AI services are subscription‑based, turning CapEx into OpEx.


• The real danger lies in operational debt—committing to tools that never deliver value.

The prudent approach is to treat AI spend as a series of experiments, each with a clear success metric before moving to the next round of investment.

• --

The Road Ahead: Practical Steps for SMB Leaders

1. Audit your data – Identify clean, structured datasets that can feed AI models.
2. Pick a pilot – Choose a high‑impact area (e.g., sales forecasting) and a reputable SaaS AI vendor.
3. Set measurable goals – Define KPIs such as % reduction in forecast error or time saved on routine tasks.
4. Build human‑in‑the‑loop processes – Ensure staff review AI recommendations before final decisions.
5. Iterate and expand – Once the pilot shows ROI, add complementary AI services (e.g., chat‑bots, document analysis).

• --

Conclusion

Big corporations may be the loudest voices in the AI arena, but their missteps are a goldmine for SMBs. By emulating the strategic uses of AI—analytics assistance, diversified toolsets, and revenue‑linked pilots—while steering clear of trend‑chasing, data‑security lapses, and over‑reliance on single vendors, small firms can reap disproportionate benefits without the billion‑dollar price tag.

For more insights from Gene Marks, visit his Forbes column and the Guardian article referenced below.

• --

References

• Marks, Gene. “Big business has shown small firms what to do – and what not to do – with AI.” The Guardian, 30 Aug 2026. https://www.theguardian.com/technology/2026/aug/30/ai-small-business
• JPMorgan Institute. “Understanding the Use of AI Among Small Businesses.” https://www.jpmorganchase.com/institute/all-topics/business-growth-and-entrepreneurship/understanding-ai-use-by-small-businesses
• Marks, Gene. Forbes Columnist. https://www.forbes.com/sites/quickerbettertech
• Reddit discussion on the Guardian article. https://www.reddit.com/r/GUARDIANauto/comments/1w2istd/tech_big_business_has_shown_small_firms_what_to