In a rapidly evolving security landscape, U.S. border security agencies are facing pressing challenges in managing burgeoning data volumes while countering increasingly complex threats.
This complexity is matched by public demands for transparency and accountability in security operations.
Zach Beus, the national security lead at i2 Group, a Harris Computer company, emphasises the crucial role of artificial intelligence (AI), data integration, and unified analytic standards in shifting these operations from reactive to proactive strategies.
Breaking down silos in border security
Beus advocates for a shift from these isolated databases to integrated platforms
Beus, drawing from his extensive experience as a former intelligence officer for the National Geospatial-Intelligence Agency and his work in various regions including the U.S.-Mexico border, highlights that the core challenge for border agencies is linked to the fragmented nature of available data.
"It's not just that the data is large," Beus shared in a BizTechReports vidcast, "It's diverse, fast-moving, fractured, and siloed. Agencies have more information than ever before, but they can't always use it effectively."
U.S. agencies like Customs and Border Protection (CBP) and Immigration and Customs Enforcement (ICE) collect vast amounts of data, from shipping manifests to social media activities. However, much of this information is locked within disparate systems, hindering efficient analysis.
Beus advocates for a shift from these isolated databases to integrated platforms capable of unifying structured, semi-structured, and unstructured data sources, which could enhance analytic efficiency through federated search capabilities across jurisdictions.
AI's transformative potential
Surveys corroborate the need for eliminating data silos. A 2025 Gartner study identified legacy systems and data fragmentation as significant barriers to leveraging AI's full capabilities in national security contexts. Successful adoption, analysts suggest, relies not just on new tools but on robust integration and governance frameworks.
Moreover, the integration of AI promises to revolutionise the role of analysts within border operations. According to Beus, AI can automate time-consuming tasks like entity resolution and pattern detection, enabling analysts to focus on interpreting complex intelligence insights.
"AI changes the role of an analyst from finding a needle in a haystack to interpreting why that needle matters," Beus noted, shifting the focus to questions of intent, context, and broader implications.
Reflecting this shift, the ISC² AI Pulse Survey (2025) reported that 70% of security professionals using AI-related tools observed improved team efficiency, attributing this to AI's ability to automate repetitive data processing, allowing personnel to engage in higher-value analytical work.
Establishing standards for AI use
Beus cautions that moving forward without established standards could pose risks
Despite these advancements, Beus cautions that moving forward without established standards could pose risks. A lack of common protocols in AI deployment may lead to legal and operational setbacks.
"At some point, whether you're a local police department or the CIA, you may be asked in a court of law how you derived information from AI," Beus explained, highlighting the necessity for consistent approaches.
This viewpoint is echoed by the Cloud Security Alliance’s survey, which found that while many in the IT and security fields anticipate AI will substantially enhance threat detection, transparency and standardisation are key to maintaining trust and accountability.
Balancing security and civil liberties
The integration of AI in border security isn't without ethical and policy challenges. The mission necessitates a delicate balance between national security and civil liberties.
Beus acknowledges that while AI's maturity brings opportunities for increased transparency and accountability, the government must ensure that oversight frameworks are in place to uphold ethical standards.
Addressing these concerns allows agencies to focus on reallocating resources more efficiently. AI applications can significantly decrease the workload, enabling more analysts to concentrate on strategic assessments previously hindered by resource constraints, thus supporting more innovative policy developments.
Advanced tools for analysts
i2 Group is advancing its Analyst’s Notebook platform to better equip defence, law enforcement, and intelligence agencies. New features include:
- Natural Language Processing: Facilitates the extraction and analysis of data from unstructured documents.
- Automated Insights: Enables rapid normalisation and visualisation of data from spreadsheets, providing instant insights.
- Geospatial Mapping: Offers real-time visualisation of movements, enhancing predictive analysis.
As Beus states, the goal is to minimise the data management burden, allowing analysts to prioritise analysis and decision-making.
Partnering for future success
Effective border security solutions demand robust partnerships that merge cutting-edge technologies with comprehensive legacy system integration. Beus emphasises the necessity for collaborative ecosystems that address the complex demands of modern border operations.
With the technological landscape continuously advancing, intelligence-led approaches are becoming increasingly crucial. Beus concludes, "We're in a really good position right now to leverage both people and technologies; we just have to put the right emphasis at the right time to support analysts on the front lines."
Discover how AI, biometrics, and analytics are transforming casino security
U.S. border security agencies are under pressure to adapt to a new intelligence reality: an environment where data volumes are exploding, threats are increasingly sophisticated, and public expectations demand both security and accountability.
According to Zach Beus, national security lead at i2 Group, a Harris Computer company, the solution lies in harnessing artificial intelligence, data integration, and shared analytic standards to shift from reactive to proactive operations.
Beus, a former intelligence officer with the National Geospatial-Intelligence Agency who has supported missions in Afghanistan, Latin America and on the U.S.-Mexico border, said the challenge facing border authorities is not simply one of size.
“It’s not just that the data is large,” he told BizTechReports in a recent vidcast interview. “It’s that it’s diverse, fast-moving, fractured, and siloed. Agencies have more information than ever before, but they can’t always use it effectively.”
Breaking down silos
Customs and Border Protection (CBP), Immigration and Customs Enforcement (ICE), and other agencies collect shipping manifests, travel records, sensor feeds, financial intelligence, and even social media data. But much of that information remains trapped in isolated databases, slowing analysis.
“We need to start shifting from siloed databases toward platforms that more easily integrate structured, semi-structured and unstructured sources of data,” Beus said. He added that federated search capabilities could help analysts conduct a single query across multiple jurisdictions, dramatically improving efficiency.
“When I was an analyst, I’d sometimes have to perform searches 16 times across different systems on the same person,” he said. “That’s not sustainable.”
Independent surveys
Independent surveys underscore the urgency of breaking down silos. A 2025 Gartner study on government productivity and AI warned that legacy systems and fragmented data remain the biggest obstacles to unlocking the full potential of artificial intelligence in public sector missions — including intelligence and other national security applications.
Analysts concluded that integration and governance, not just new tools, are essential for success.
Role of AI
Beus argued that artificial intelligence can redefine how analysts work. Machine learning and natural language processing (NLP) can automate tasks like entity resolution, link analysis, and pattern detection, allowing humans to concentrate on higher-level assessments.
“AI changes the role of an analyst from finding a needle in a haystack to interpreting why that needle matters,” he said. “Instead of just answering who, what, when and where, analysts can now focus on the why — intent, context and long-term implications.”
Industry data reflects this shift. The ISC² AI Pulse Survey (2025) found that 70 percent of security professionals using AI-enabled tools reported improved team effectiveness, with most saying the technology freed them from repetitive data sorting and let them focus on higher-value tasks.
Establishing standards
Still, Beus cautioned against moving too fast without establishing standards. A lack of common protocols for using AI could create legal and operational risks.
“At some point, whether you’re a local police department or the CIA, you may be asked in a court of law how you derived information from AI,” he said. “If there’s not a common approach, we might get into some really big problems.”
That concern is widely shared. The Cloud Security Alliance’s “State of AI and Security” survey (2024) found that while 63 percent of IT and security pioneers expect AI to significantly enhance threat detection, many stressed the need for transparency and standardisation to maintain trust and accountability.
Security, civil liberties, and the workforce
Indeed, trust will play a key role in encouraging acceptance and adoption of AI in border operations, because the mission itself raises ethical and policy considerations. “The government will need to balance national security and civil liberties,” Beus said.
The good news is that growing maturity around how AI applications are used is being accompanied by new oversight measures designed to ensure transparency and accountability. Agencies are beginning to adopt explainability frameworks and audit practices that help demonstrate how algorithms arrive at their findings, reinforcing ethical utilisation in sensitive missions.
Reallocating scarce resources
As those concerns are addressed, agencies can focus on how AI can reduce workloads and reallocate scarce resources. “For instance, it might have taken a hundred analysts to sift through or scan manifest logs, but now it takes 10 analysts,” he explained.
“So now we have 90 analysts that can do other things. They can look for long-term intelligence challenges. They can conduct strategic assessments that previously weren’t possible because we didn’t have the human capital to do it. Now we can be much more cutting edge — and from a policy standpoint, that’s a significant change.”
Tools for the analyst community
i2 Group, which has served defence, law enforcement and intelligence organisations for more than 30 years, is updating its flagship Analyst’s Notebook platform to meet these challenges. New features include:
- Natural Language Processing: Analysts can drag and drop unstructured documents, and the system automatically extracts entities, links and properties.
- Automated Insights: Data from spreadsheets can be normalised and visualised instantly, producing dashboards that highlight critical connections.
- Geospatial Mapping: Movement of phones, vehicles or people can be visualised in near real time, providing predictive insights into trafficking or migration patterns.
“Our focus is making the analyst less of a data manager and more of an analyst,” Beus said. “We want to give them tools that simplify visualisation, collaboration and sharing — whether that’s through digital files, PowerPoint decks, or wall-sized maps for operational planning.”
Partnerships as an imperative
Beus stressed that border security has no one-size-fits-all solution. Agencies need ecosystems of partnerships that blend cutting-edge tools with legacy integration expertise. “Their missions are incredibly complex,” he said.
“The most effective partnerships are those that combine AI and graph analytics with the ability to connect old and new technologies under real-world constraints.”
Intelligence-led methods
As border management grows more high-profile and technologically intensive, Beus sees intelligence-led methods as essential.
“We’re in a really good position right now to leverage both people and technologies,” he concluded. “We just have to put the right emphasis at the right time to support the analysts on the front lines.”
