THE APEX TIMES
Survey by Booz Allen Finds Federal Agencies Accelerating Agentic AI, but Security and Trust Lag
More than half of federal IT and cybersecurity decision makers say their agencies are already deploying or piloting AI systems that can act autonomously, but fewer than one-third express high confidence they can do so securely, according to a new Booz Allen survey.
Federal agencies are moving quickly toward “agentic” artificial intelligence, systems designed not just to generate responses but to take actions, yet a sizable trust and security gap remains, according to a new survey released by Booz Allen Hamilton.
The survey, published July 21, is featured in the latest issue of Velocity, Booz Allen’s ongoing publication focused on emerging technology and mission innovation. It examines what government cybersecurity leaders see as opportunities and risks as autonomous AI enters IT and cyber operations.
Among more than 100 federal IT and cybersecurity decision makers surveyed, 58% said their agencies have already deployed or are piloting AI agents. However, only 28% said they have high confidence in their ability to deploy those systems securely, suggesting a widening divide between adoption plans and governance readiness.
Booz Allen said the survey also points to continued concerns about adversaries using AI-enabled cyber capabilities. Seventy-nine percent of respondents reported they are “very” or “extremely concerned” that opponents could use AI to accelerate cyberattacks against their agencies over the next 12 to 18 months.
To address the concerns, Booz Allen said the Velocity edition lays out a systems-level approach to deploying agentic AI securely, spanning the full AI lifecycle rather than treating security as a one-time checklist. The report emphasizes identity governance, continuous monitoring, red teaming, formal validation, and the use of zero trust principles.
Identity governance refers to managing who or what is allowed to access systems and data, including permissions for AI-driven workflows. Continuous monitoring involves tracking performance and behavior after deployment, while red teaming and formal validation are designed to test an AI system against adversarial scenarios and to confirm it meets defined requirements. Zero trust principles shift security away from implicit trust inside networks and toward ongoing verification.
Booz Allen said the survey was conducted in partnership with Market Connections and involved an online survey of 105 federal government decision makers and influencers involved in IT and cybersecurity strategy. The fieldwork was conducted in April 2026.
For Booz Allen, the survey is positioned as both a warning and a planning tool for organizations wrestling with how to gain operational benefits from agentic AI without losing control of what those systems do and how they can be held accountable. The company described the underlying theme as a challenge for cybersecurity leaders: harnessing speed and autonomy while ensuring security and accountability.
Still, the release does not detail what specific agentic AI systems are being piloted, how agencies define “high confidence,” or what security gaps respondents are most concerned about. It also does not break out results by agency type, mission area, or procurement status, limiting how far the findings can be used to pinpoint where implementation risk is concentrated.
Looking ahead, the key question for federal adopters will be whether security measures such as identity governance and continuous monitoring keep pace with expanding deployments. As agencies push agentic tools from pilot into production, Booz Allen’s survey suggests the deciding factor may be governance maturity, including validated controls, testing discipline, and ongoing verification of AI behavior in operational environments.
Why It Matters
- The gap between deployment and security confidence suggests agencies may need faster progress on governance, testing, and operational controls to match adoption timelines.
- High concern about AI-accelerated cyber threats indicates pressure to treat agentic AI as both a defensive opportunity and a risk factor in threat models.
- If agencies proceed with autonomy, demand may rise for security frameworks that can demonstrate accountability and validated performance rather than relying on vendor promises.
Sources
Key Facts
- Booz Allen released a survey on agentic AI adoption in federal IT and cybersecurity environments.
- 58% of 105 federal IT and cybersecurity decision makers surveyed said their agencies have deployed or are piloting AI agents.
- 28% said they have high confidence in deploying AI agents securely.
- 79% of respondents reported being “very” or “extremely” concerned about AI-enabled adversaries accelerating cyberattacks in the next 12 to 18 months.
- The survey was fielded in April 2026 and conducted with Market Connections.
- Booz Allen said its accompanying Velocity edition highlights security controls across the agentic AI lifecycle, including identity governance, continuous monitoring, red teaming, formal validation, and zero trust.
Defense Related
Report: Exxon Mobil joins bidders for Shell’s U.S. chemicals assets, a potential shift for XOM’s refining-and-chemicals outlook
Exxon Mobil Holdings has reportedly entered the race for Shell’s U.S. chemicals business, an asset package that includes four plants across Louisiana, Texas and Pennsylvania. The bid, if it proceeds, could change how investors think about XOM’s downstream growth and capital allocation.
UPS says its reorganization will lean more heavily on global logistics than domestic parcel operations
The shipping company outlined a plan to restructure operations around new global standards, framing the change as a way to strengthen cross-border capabilities while maintaining its parcel network.
General Dynamics shares fall more than the broader market in late-session trading
General Dynamics (GD) closed at $371.35 on Aug. 31, down 2.1% versus the prior trading day, according to Yahoo Finance.
Eli Lilly to buy Merdia Biosciences in a deal valued at up to $2.88 billion, indicating renewed focus on pipeline expansion
The acquisition, reported as worth as much as $2.88 billion, adds another chapter to Lilly’s ongoing buy-or-build approach as biotech rivals also compete for late-stage assets and platform-like capabilities.
Nvidia hardware momentum meets a new choke point: copper, not cash, HIVE Digital’s Frank Holmes says
A Wall Street executive argues that today’s AI funding is not the limiting factor. The bottleneck, he says, is the physical supply chain behind data centers, where power and copper wiring needs can outstrip available materials.
Broadcom’s Sept. 2 earnings set up a high-stakes test for its AI narrative
Ahead of its next quarterly report, Broadcom is drawing attention from investors who are trying to separate short-term uncertainty from longer-term demand linked to artificial intelligence.
Palantir CEO Alex Karp pushes back on “tokenmaxxing,” pitching real-world AI value over hype
In comments highlighted by Yahoo Finance, Palantir’s CEO argues that investors should separate durable, use-case-driven AI progress from speculative “token industrial complex” narratives.
FTC and 22 states sue Amazon, alleging inflated prices in online ads scheme
The Federal Trade Commission and a coalition of states filed a lawsuit accusing Amazon of misleading advertising customers and defrauding them through inflated ad pricing. Amazon has not been found liable, and the company’s response was not included in the announcement referenced by the reporting.
Tesla shares rise after investors refocus on long-term autonomous driving potential
Tesla (TSLA) gained about 4.9% in the afternoon session, according to market coverage, as traders appeared to anchor on the company’s longer-term self-driving ambitions.
Tim Cook’s final day as Apple CEO caps a 15-year push into services, wearables and payments
Apple marks the end of Tim Cook’s tenure as chief executive, a period defined by new hardware categories and a growing reliance on services, culminating in a market value described in a recent report as topping $4 trillion.
