As adoption of artificial intelligence accelerates across government and public sector markets, companies most likely to build lasting businesses treat trust as foundational. Early investments in transparency, buyer education, and independent validation are necessary to produce compounding returns in markets where credibility is everything.

For most of the software industry’s history, the central question a buyer asked was simple: Does this product work? Companies that built reliable, effective software won customers. The relationship between a vendor and its customer was straightforward. The vendor built a tool, the customer used it, and data stayed in predictable places and did predictable things.

AI has changed that calculus completely. The barrier to building capable software has been lowered. Capable AI is no longer the exclusive domain of well-resourced technology companies.

What has become harder to establish is trust.

WHY TRUST IS LAGGING

Currently, in the AI sector, capability and confidence are moving in opposite directions. According to Stack Overflow’s 2025 Developer Survey, AI adoption among developers climbed to 84%. However, 46% of developers report actively distrusting the accuracy of AI output.

Buyers today are asking a different set of questions before they commit to an AI vendor. What happens to our data once it enters your system? Is it being used to train your models? Could proprietary or sensitive information find its way into outputs that others can access?

The questions above matter across industries. But they carry particular weight when the buyer is a government agency, a law enforcement department, or a public sector organization. As someone who has spent over a decade investing in AI companies that build products for exactly these environments, I’ve watched trust become the variable that separates companies that grow sustainably from those that stall out despite strong technology.

The scale of what’s at stake is significant. According to a Brookings Institution analysis, federal agencies documented more than 3,600 individual AI use cases in 2025 alone—five times the number reported just two years earlier. The agencies and institutions adopting these tools are accountable to the public in ways that private sector buyers are not. When an AI system deployed in a public safety context produces a flawed output, the consequences can affect people’s lives in profound ways. This, in turn, shapes how legislators think about AI regulation, how other agencies approach procurement, and how the public perceives the technology as a whole.

The 2025 Edelman Trust Barometer found that only 32% of Americans currently trust AI. What separates the AI companies that earn trust from those that lose it often comes down to how they handle the questions buyers are already asking.

Many companies underestimate how early in the evaluation process a buyer forms their impression. Vague answers about data handling, overly technical language that obscures rather than clarifies, and an absence of clear documentation about how a system reaches its outputs all create additional hesitation. Companies that proactively address these questions tend to move through procurement more smoothly and build relationships that hold up over time.

WHAT’S NEEDED TO GAIN TRUST

Investing in buyer education is one of the more underrated decisions an AI company can make. In high-responsibility environments, a buyer who genuinely understands how the product works, how it reaches its outputs, and where its limitations lie is far more likely to deploy it effectively and remain a long-term customer.

Verifiable signals matter far more than stated ones. Simply telling a government buyer that your system is secure and your data practices are sound carries little weight on its own. Third-party audits, independent technical validations, and customers willing to speak publicly about their experience provide the kind of evidence that buyers need to defend decisions to their stakeholders.

As AI capabilities continue to proliferate across the public sector, companies building for government and law enforcement clients will find that the technology itself becomes a smaller part of what differentiates them. The field will have capable products. What it will have less of is companies that have earned the kind of credibility that allows high-stakes institutions to adopt their AI tools confidently, defend those decisions publicly, and build on them over time. That is where the durable position gets built. For AI companies operating in these markets, it may be the most important investment they make.

About Elliott Broidy

Elliott Broidy is Chairman and CEO of Broidy Capital Holdings, as well as Co-Founder and CEO of LEO Technologies and Chairman of Cynapse.ai.