GDIT's Gilliland: Humans Will Master AI, Not Be Replaced By It
WASHINGTON - As skepticism grows about artificial intelligence’s real-world benefits, General Dynamics Information Technology is betting on a different approach: empowering employees to master AI tools rather than viewing them as replacements.
Amy Gilliland, president of GDIT, articulated this philosophy during a media roundtable Tuesday, emphasizing that the company’s AI investment strategy prioritizes education and internal capability development over immediate return on investment metrics.
“It’s early days,” Gilliland told reporters. “Right now, the battle that we are waging is education: It is about understanding what AI can do for you.”
Her comments come amid industry-wide questions about AI’s tangible value. A recent MIT study found that 95 percent of surveyed companies had achieved zero return on generative AI investments as of June. Gilliland countered that applying traditional ROI frameworks to early-stage AI adoption is premature. Instead, GDIT is focusing on tailoring large language models to company-specific environments and training employees on practical applications.
The numbers reflect this commitment: GDIT reports taking 10 times as many AI training courses in 2025 compared to 2024. Employees are particularly enthusiastic about using AI for routine coding tasks, allowing human engineers to focus on intellectually challenging work rather than repetitive code generation.
Beyond internal operations, GDIT sees significant potential for LLMs in proposal drafting. With over 1,000 annual bids, AI tools can perform initial document analysis and identify relevant past projects, accelerating the proposal process while humans retain final review authority.
GILT also implemented an AI-powered internal mobility tool that matches employees with new opportunities—both permanent position changes and short-term “gigs” lasting six days to six weeks. This platform was highlighted as a key application for GDIT’s newly unveiled Mission Emerge Center.
Located in Springfield, Virginia, just five minutes from the National Geospatial Intelligence Agency campus, the 5,000-square-foot center serves as a showroom and testing ground for emerging AI and geospatial intelligence products. The facility demonstrates surveillance video analysis, 3D mapping, and specialized chatbots in unclassified sandboxes, allowing government customers to evaluate technology before committing to lengthy certification processes.
Gilliland emphasized that this innovation hub supports a broader military imperative: accelerated technology adoption. The traditional defense acquisition timeline—measured in months or years—is becoming obsolete. Ukraine’s experience demonstrates the military value of rapid technology adaptation in real time.
The DoD increasingly expects agile development cycles with sprint reviews every six weeks rather than annual updates. Gilliland, a Naval Academy graduate and former Navy officer, illustrated the shift with a naval analogy: historically, ships received software updates only during port deployments, sometimes 18-27 months apart. Modern requirements demand continuous, underway upgrades.
“That is where we’re headed, and I think that’s where we need to be,” Gilliland said, aligning GDIT’s AI strategy with defense industry modernization imperatives.
Source ID: SRCE-2025-1764795886248-1131