10

Loading video...

Frank Sahlein

By Frank Sahlein

0

Share

Newsweek is a Trust Project member

See more of our trusted coverage when you search.Prefer Newsweek on Googleto see more of our trusted coverage when you search.


American businesses are adopting artificial intelligence faster than they are learning to protect what they feed into it, I believe. That gap is the defining risk of this moment, and it seems to me that most companies are not prepared for it.

The pressure to adopt is real. According to Thryv's 2026 AI and Small Business Adoption Report, 66 percent of small businesses now use AI, up from 55 percent the previous year. Adoption is no longer the question. The question is what happens to the information employees put into these tools.

The answer is uncomfortable. As per HRD America, reporting on a 2026 Wakefield Research survey for PagerDuty, 43 percent of office professionals had entered work-related correspondence into public AI tools outside company systems. Roughly a third had shared customer or financial data with those same tools. I'm not saying these are rogue actors.

Read More on Leadership Insights

I'd say the vast majority are just employees trying to do their jobs faster, using the tools available to them, often without understanding where the data goes or how it is retained.

According to GTIA's 2026 research, 24 percent of small and midsize businesses cite data security, privacy, and compliance concerns as a top deployment barrier to AI. The concern is legitimate. But avoiding AI is not a strategy. I believe the businesses that will lead the next decade are not the ones that refuse to adopt. They are the ones that build environments where adoption does not require sacrificing control.

I have spent years watching industries struggle with this exact problem. The child activity center industry, where I built my first business, is no different from any other. Operators want the efficiency AI promises. They do not want to risk handing over their proprietary training materials, customer records, or strategic plans to a platform that could use that information to train models they do not control. The fear is not paranoia. It is rational.

In my experience, the mistake most businesses make is treating AI adoption as a binary choice: use the public tools and accept the risk, or stay out of the game entirely. That framing is wrong. The real choice, I would suggest, is between using tools designed for general consumers and using environments built for specific industries with specific privacy requirements. The innovation behind MetaSpheres is designed to help industry experts monetize AI for themselves safely, while keeping their IP protected.

A private, industry-specific AI environment changes the calculus. The platform we built connects a business's own information, its customer systems, its training materials, and specialist industry content in one place, within a system that is designed not to export proprietary data. The technology is virtually the same, I would argue. The architecture is different. And the architecture is what determines whether a business can use AI without becoming a cautionary tale.

Avoid Compounding the Issue

The stakes are not abstract. If businesses continue to rely on public tools without governance, the consequences will compound. Employees are likely to keep sharing sensitive information. There is a risk of competitors gaining access to insights that should have remained internal. And the businesses that took the risk could spend years trying to recover a competitive advantage they gave away by accident.

The alternative is achievable. If businesses adopt AI through environments that prioritize privacy and control, they should be able to move faster without exposing what makes them valuable. They can give employees access to intelligent tools without wondering what happens to the data those tools consume. They should also be able to connect with industry experts and suppliers without handing over their customer lists.

The platform we built is one such approach. There will be others. What matters is that businesses begin evaluating AI environments on privacy, control, and industry relevance rather than on the novelty of the tools themselves.

This is not a call to slow down. It is a call to build the right infrastructure before the bad habits become embedded. I believe that the businesses that treat AI governance as an afterthought will spend the next few years managing the consequences. The ones that treat it as a foundation will spend those years building an advantage.

AI is not going away. The question is whether the businesses using it will remain in control of what they know. That question will be answered by the choices made now, not by the tools themselves.

Frank Sahlein is co-founder of MetaSpheres, a private AI platform, and all opinions are the author’s own.

View 0 Comments

Request Reprint & Licensing

Submit Correction

View Editorial & AI Guidelines