Corporate commentary on artificial intelligence often resolves into one of two flavors: breathless enthusiasm or cautionary hand-wringing. Margarita Howard, the founder and CEO of government contractor HX5, is trying to offer something more useful.
She has publicly committed to aggressive AI adoption while also flagging, with some specificity, the ways the technology could damage a business like hers. Both positions, she treats as obligations.
“The recent growth of artificial intelligence is certainly something that is exciting to see and think about,” Howard said. “But while recognizing and appreciating the immense power and potential of AI, it’s equally important that we always remain cognizant of the potential dangers it may present and how it could be used to impact our business, for the better or worse.”
The Adoption Commitment
Howard’s position on adoption is unambiguous. “If you don’t embrace AI, you’re just going to be gone,” she said. “We’re adopting it heavily.”
“We’re actually developing AI tools internally that we’re using and seeing benefits from,” she continued.
The potential value of AI-related federal contracts grew from $355 million to $4.6 billion in a single year, between August 2022 and August 2023, according to a Brookings Institution analysis. The jump was driven substantially by Pentagon programs. The July 2025 DoD awards of $200 million each to Anthropic, Google, OpenAI, and xAI for agentic AI development confirmed that federal AI investment is a sustained commitment, not a one-cycle budget line.
Howard also looks forward to what AI will mean for the contractor-government relationship itself.
“We believe that in the years to come, and certainly by 2035, government agencies will increasingly utilize AI to streamline procurement processes, evaluate contractor performance, and probably predict future needs based on historical data that they collect,” she said. “Compliance protocols will be automated. Contractors will be required to integrate systems that provide continuous reporting and real-time audit capabilities.”
What AI Does for a Government Contracting Business
At a practical level, Howard describes productivity as the immediate benefit. Automating data collection and analysis frees engineers and program managers for work that requires human judgment, a real value proposition for a firm whose workforce is concentrated in high-cost STEM professionals on billable contracts.
AI also supports proposal development, where analyzing solicitation requirements, aligning technical responses to evaluation criteria, and checking for compliance gaps are exactly the kind of structured, document-intensive tasks that AI tools handle well.
There is also a workforce dimension. HX5 has modernized internal communication systems to meet younger employees on platforms they use: messaging tools, digital project management. AI-integrated workflows have the potential to fit naturally into that shift and help close the productivity gap between experienced staff and newer hires who are still developing government program fluency.
The Dangers
Of course, the dangers of AI remain real, particularly in sensitive industries involved in government contracts.
CISA released guidance in May 2025 specifically addressing AI data security risks for organizations including the Defense Industrial Base, identifying data supply chain risks, maliciously modified training data, and data drift as three distinct threat categories.
Meanwhile, the FY2026 National Defense Authorization Act directed the DoD to build an AI security framework into DFARS and the CMMC program, extending mandatory AI security requirements across the defense industrial base. The regulatory framework is still being built, which creates compliance ambiguity, another risk in itself.
Holding Both Positions
What makes Howard’s stance on AI practically useful, rather than merely cautious, is that she doesn’t treat the risks as arguments against adoption. She treats them as conditions that responsible adoption must account for. The danger isn’t the technology itself, but rather moving too fast without operational controls around it, or moving slowly while competitors establish AI capabilities that start appearing in source selection evaluations.
For a mid-size contractor operating on cost-sensitive government programs, both failure modes are real. HX5’s approach — internal development, external consultants, ongoing vigilance about what AI tools touch and where data goes — reflects a contractor who has internalized that compliance isn’t a constraint on the business. It is the business.
