October 2025 Summaries

Government agencies face the challenge of integrating AI into their services while maintaining high security standards, a topic explored during the Box Public Sector Virtual Summit. This event emphasized the need for a secure, unified content platform that allows agencies to modernize their workflows without compromising compliance and citizen trust. Strategies discussed include building secure foundations, deploying AI agents with human oversight, ensuring compliance without sacrificing speed, customizing AI solutions to suit specific agency needs, and integrating protection and trust into every aspect of AI deployment. Practical applications and successful case studies from various government sectors, such as secure evidence repositories and automated contract analysis, illustrate how targeted AI implementations can enhance government processes. The summit underscored the importance of balancing innovation with risk management to improve citizen services while maintaining transparency and security.

Mercer Advisors, an independent fiduciary financial advisory firm, has significantly enhanced its efficiency and client service by integrating AI-powered document workflows with Box, transforming its traditionally manual processes. This strategic partnership has allowed the firm to automate document handling, leading to a fivefold increase in data extraction speed and reducing the time to generate quotes from two weeks to five minutes. This automation not only streamlines client onboarding but also enables advisors to focus on client relationships by reducing administrative burdens. President Daniel Gorvitch emphasizes that this technology amplifies human expertise rather than replacing it, resulting in productivity gains of 30-50% for financial planners. The initiative has improved staff satisfaction and confidence by minimizing error risks and enhancing the analytical focus of their roles. Mercer Advisors plans to further expand its AI capabilities to fully automate client lifecycle management while maintaining compliance standards. This case illustrates the potential of intelligent automation to scale personalized service in the financial advisory sector, enhancing both efficiency and client satisfaction without sacrificing the personalized care essential to their fiduciary responsibility.

The new integration of the Box MCP server with the GitHub MCP registry provides developers with a secure and seamless connection between enterprise content stored in Box and AI development platforms like GitHub. This collaboration allows developers to access Box tools within their development environment, significantly improving workflow efficiency and security by providing a live connection to essential business documents such as design documents, product specifications, and internal knowledge. The Box MCP server acts as a secure bridge that enforces Box's native permissions and governance rules, enabling AI applications to perform advanced functions like search and metadata extraction without custom integrations. This setup empowers developers to enhance their AI-powered workflows by grounding them in relevant business content while leveraging GitHub Copilot to facilitate onboarding, context-aware code generation, troubleshooting, and documentation automation. The integration marks a significant step toward creating an open AI ecosystem, making it easier for users to build smarter, enterprise-grade AI applications while maintaining productivity and reducing the need for context switching.

Box has achieved the PROTECTED authorization level through the Information Security Registered Assessors Program (IRAP) assessment, meeting the Australian Government's stringent security standards for handling highly sensitive information. This accomplishment underscores Box's dedication to providing robust security controls, such as advanced encryption and access management, which are critical for protecting government data against significant threats. Serving over 10,000 government agencies worldwide, Box's Intelligent Content Management platform facilitates collaboration, manages content lifecycle, and enhances business workflows with enterprise-grade AI, while ensuring compliance and security. The IRAP assessment not only validates Box's capabilities in securing sensitive government information but also offers public sector agencies streamlined procurement processes, reduced audit burdens, and improved operational and compliance risk management, thereby facilitating the secure modernization of content management.

Box has announced a major update to Box Drive for Windows, enhancing its integration and functionality within Windows environments, particularly for hybrid work models and diverse hardware ecosystems. The update introduces support for Arm-based Windows devices, allowing Box Drive to run natively on these devices and facilitating seamless content management across more efficient, always-connected hardware. Known folder redirection is now available, enabling automatic synchronization of local folders like Documents and Desktop to Box Drive, enhancing data protection and compliance while improving user onboarding and device management. Additionally, the update provides improved offline access control, allowing users to mark specific files for offline availability and manage downloaded content more effectively in environments with limited connectivity. Other improvements include flexible storage management, enhanced user experience with progress indicators in File Explorer, and better visual integration with consistent iconography. These enhancements aim to make Box Drive a more integral part of the Windows operating system, offering a more stable and reliable user experience.

A major update to the Box MCP server community edition has introduced new features that make content sharing more efficient, particularly through enhanced support for Box shared links and web links. This update includes 16 new tools that allow users to generate, manage, and modify secure URLs for files, folders, and web links, facilitating direct, read-only access to resources with varying levels of access control. The integration of these tools with conversational AI, such as Claude desktop, enables users to perform complex content management tasks through natural language commands, streamlining workflows and lowering barriers to adoption. This innovative approach enhances accessibility for users of all technical levels and offers extensive integration possibilities with other MCP tools, fostering more automated and efficient workflows. The update signifies the transformative potential of MCP integrations, and Box is committed to further expanding these capabilities based on community feedback, encouraging users to experiment with and contribute to the ongoing development of its ecosystem.

The partnership between Box and Tata Consultancy Services (TCS) aims to harness AI to transform unstructured content into a competitive advantage for businesses by delivering industry-specific solutions that address real customer challenges, enhance productivity, and facilitate digital transformation. With the integration of Box's AI-first content platform and TCS's industry depth and extensive network, the collaboration will provide scalable and intelligent solutions across various sectors such as financial services, healthcare, manufacturing, retail, and public sector. The partnership is expected to automate routine processes, reduce risks, speed up decision-making, and extract strategic insights, thereby boosting productivity and driving business outcomes. By combining Box's technology with TCS's expertise, the collaboration seeks to unlock new growth opportunities for joint customers worldwide through end-to-end solutions that streamline operations and foster innovation.

The Virginia Department of Health Professions (DHP) has significantly enhanced its operations by adopting Box for Intelligent Content Management, addressing the inefficiencies and security challenges of its previous paper-heavy, manual processes. Box allows DHP to securely store, manage, and share sensitive documents, aligning the workforce around a centralized content platform that supports compliance with stringent security standards set by Virginia’s Commonwealth Security Group. By leveraging Box’s unlimited storage and advanced security features such as Box Shield, DHP has improved collaboration, reduced IT support calls, and sped up licensing and investigation processes. The agency is now poised to explore AI advancements for tasks like case file summarization and statistical analysis, aiming to harness the potential of unstructured data while maintaining robust security measures. This transition to a digital workflow not only safeguards sensitive information but also enhances customer satisfaction by providing easy access through multiple platforms like Box Drive, Mobile, and Web, ultimately transforming cumbersome regulatory processes into efficient digital operations.

Software development is at the forefront of the AI transformation, with engineers utilizing their unique laboratory-like conditions to innovate rapidly and increase productivity. This environment, characterized by immediate feedback, ruthless tool selection, and constant innovation, has allowed early adopters to experience significant productivity gains, such as a 37% increase. Engineers, who are both builders and users, benefit from digital workflows that facilitate quick assessments and adaptations, resulting in an ongoing improvement cycle. The key to leveraging AI effectively lies in designing AI-native systems, rather than merely adding AI onto existing frameworks, as this approach fosters meaningful productivity improvements. As AI reshapes coding workflows, its influence is expected to extend to other domains like marketing, legal, and customer support, where AI can automate processes and enhance execution speed. Successful implementation requires mastering skills like effective prompting and output review, focusing on designing precise tasks and acceptance criteria, and emphasizing breakthrough changes over incremental improvements. The transformation observed in software engineering is set to influence broader knowledge industries, demonstrating that AI agents have changed software development fundamentally and for the better.

Box SDKs have undergone significant updates to streamline integration with the Box API, focusing on enhancing the developer experience through a new major version (v10) that consolidates previously separate SDKs into a unified package. This latest version, which marks a shift from manually maintained packages to generated ones, introduces breaking changes and improves API support, updates, documentation, and convenience methods. Developers are encouraged to migrate to v10 and beyond to benefit from ongoing support and new features, as the Box Next Generation SDKs are no longer supported as separate entities. Detailed migration guides are available to assist with the transition, ensuring that existing applications can continue to function without disruption while taking advantage of the new consolidated architecture. By 2027, support will continue for the consolidated versions, and developers are advised to use v10 or higher for new projects to ensure they receive security patches and improvements.

Modernizing government involves reducing administrative burdens, enhancing oversight, and delivering transparent services, as emphasized by former federal CIO Jason Gray, who now works at Box. Drawing from his extensive experience across various federal agencies, Gray highlights the importance of using scalable solutions to address federal missions' complexities, compliance, and resource limitations. At Box, he observes how the platform consolidates multiple operational needs—such as workflow automation, content management, and threat detection—into a secure, auditable environment, enabling agencies to operate efficiently and safely. Emphasizing responsible AI, Box employs multiple large language models within a FedRAMP High-authorized platform, allowing federal teams to select suitable models for diverse tasks while maintaining data security and auditability. This approach aligns with the broader trend in federal technology towards consolidation and efficiency, where platforms that integrate various functionalities are preferred, ensuring quick access to authoritative information and reinforcing a "trust but verify" philosophy.

Congo Brands, a fast-growing food and beverage company, has successfully tackled the challenge of managing its vast amount of unstructured data by adopting Box for Intelligent Content Management. The company, known for collaborating with influencers like Logan Paul and Kim Kardashian, faced inefficiencies in document management and compliance risks due to scattered file storage. By leveraging Box's AI tools, Congo Brands has significantly streamlined operations, cutting contract review time and enhancing creative efficiency through metadata-powered automation. The implementation of Box allows for easy migration, improved security, and quick adoption across teams, boosting efficiency in managing over 1,300 contracts. With AI-enhanced capabilities, Congo can now automatically extract metadata from various files and images, freeing up valuable employee time for innovation and growth. As the company aims to integrate further with its product management systems and enhance retail workflows, it positions itself as an AI-first entity with a competitive edge in the consumer packaged goods market.

Box has introduced Institutional Storage, a specialized solution designed for higher education institutions to consolidate their research datasets, administrative records, and other data into a single secure repository, addressing the challenges of managing increasing content volumes with fragmented and expensive storage solutions. This system provides unlimited user-owned storage and supports compliance with regulations such as FERPA, HIPAA, and GDPR, offering enterprise-grade security and centralized governance to eliminate storage silos. Its features include automated retention, AI-powered classification, and metadata-driven organization to enhance content discoverability and usability, facilitating collaboration and focusing on teaching and research. Institutions can benefit from simple, transparent pricing with the flexibility to scale storage as needed, and additional capabilities for integrating outputs from automated pipelines and custom applications. Box’s platform also offers industry-leading admin controls and service-account models for precise access management and content differentiation, while Box AI assists in making data easier to find and utilize.

Stanford Medicine is at the forefront of integrating AI into healthcare by balancing innovation with responsibility, ensuring patient care is enhanced rather than disrupted. Key strategies include democratizing AI access across the organization, prioritizing workflow integration over technological sophistication, and embedding continuous learning and governance to adapt AI systems as they evolve. Rather than isolating AI initiatives within an elite team, Stanford empowers all staff with AI tools and education, fostering grassroots innovation across departments. This approach has led to practical applications like ambient scribes in patient visits, which improve the physician-patient interaction by automating clinical notes. Stanford distinguishes between AI's roles in automation and augmentation, focusing on enhancing human capabilities and setting realistic expectations for AI's probabilistic nature. Comprehensive governance frameworks ensure AI implementations are safe and valuable, with ethics assessments and ongoing monitoring to align with patient safety. By exploring advanced AI applications, such as using transformer models in cancer care, Stanford Medicine exemplifies a roadmap for healthcare organizations to personalize treatment decisions and navigate the AI revolution effectively.

Box is embracing AI as a capability expander, focusing on innovation and efficiency by integrating AI into its operations and fostering a culture of AI curiosity. The company applies its AI-first philosophy through five principles: capability expansion, human-AI partnership, AI-native design, privacy and security, and treating data as a strategic asset. These principles guide Box in using AI to streamline customer support, enhance sales processes, and optimize professional services, saving time and improving productivity while ensuring human oversight and decision-making. Box's journey along the AI maturity curve involves implementing AI to handle data-driven tasks and workflows while humans focus on strategic thinking and innovation. By integrating AI across various functions, Box aims to accelerate decision-making and amplify employee productivity without replacing human work.

As enterprises increasingly scale their AI deployments, the focus has shifted to achieving economic efficiency at high production volumes, particularly in areas like customer support, document processing, content operations, and automated workflows. To address this, Anthropic's Claude Haiku 4.5 model, now supported by Box AI, offers an efficiency-optimized solution that matches the performance of the more advanced Claude Sonnet 4.5 but at standard pricing. This allows organizations to handle high-volume workloads economically, making it suitable for tasks like answering documentation questions, extracting structured data, and supporting intelligent automation. By expanding the range of models in Box AI, Claude Haiku 4.5 provides flexibility in AI deployment architecture, enabling teams to use it for large-scale, routine tasks while reserving premium models for complex scenarios. Now available in Box AI, Claude Haiku 4.5 offers organizations a cost-effective way to manage AI workloads at scale.

Valmark Financial Group, a seasoned independent broker-dealer and registered investment advisor, has significantly enhanced its document management capabilities by adopting Box's Intelligent Content Management system, which employs AI to streamline processes and improve efficiency. Previously relying on a legacy on-premises system, Valmark faced challenges with speed, security, and innovation as they managed vast amounts of unstructured data, such as insurance policies and onboarding materials. By migrating to Box, Valmark has drastically increased data extraction from 4,000 to over 240,000 data points annually, reduced manual processing time by up to 48 hours on certain tasks, and enabled AI-powered search that provides instant answers to advisors. The integration of AI tools, such as Box AI Extract Agents and custom AI Studio Agents, has allowed Valmark to automate metadata extraction and improve search functionality, contributing to better job satisfaction by shifting staff focus to higher-value tasks. The collaboration with Box has not only improved operational efficiency but also positioned Valmark for continued innovation in the financial services industry, with plans to further leverage new AI features and capabilities.

The General Services Administration's OneGov initiative is facilitating the modernization of IT infrastructures in federal agencies by offering cost-effective solutions like Box's Enterprise Plus for Government and Enterprise Advanced for Government. These Intelligent Content Management solutions provide secure, scalable, and AI-powered content management capabilities, backed by FedRAMP High and IL-4 authorization, to enhance productivity and compliance. Enterprise Plus focuses on secure collaboration with features like unlimited AI queries, robust security, and native e-signatures, ideal for agencies managing critical workflows such as disaster response. Enterprise Advanced expands on these capabilities by offering no-code app building, AI automation, and extended file support, enabling rapid digital transformation and operational efficiencies. Both offerings, available at discounted rates through OneGov, support federal agencies in enhancing their mission-critical operations with advanced technology and streamlined procurement processes.

Box and Guidewire have successfully integrated their platforms to enhance the efficiency and accuracy of property and casualty insurance workflows by combining Guidewire's InsuranceSuite with Box's Intelligent Content Management system. This integration allows insurance carriers to manage submissions, claims, and litigation documents in a single secure platform, leveraging AI-driven document intelligence to automate the classification and extraction of data, thereby reducing manual entry and errors. Teams across the insurance ecosystem can access consistent content, whether through Guidewire or Box, supported by enterprise-grade security and compliance measures. As AI agents become increasingly integral to the industry, they are expected to automate document handling and data entry, allowing professionals to concentrate on risk assessment and customer satisfaction. The partnership aims to streamline core processes, accelerate FNOL processing, and reduce claim cycle times, offering insurers a pragmatic approach to achieving faster decisions, fewer errors, and improved policyholder experiences.

Box AI has integrated multilingual models from Mistral AI, specifically Medium 3 and Small 3.1, to enhance its capabilities in processing and understanding content across various languages and contexts. This integration allows enterprises to manage and analyze content in multiple languages efficiently, supporting tasks like translation, cross-border contract analysis, multilingual customer support, and international compliance. Mistral's models, known for their efficiency and advanced multilingual capabilities, enable global organizations to work seamlessly by preserving context and nuance in multilingual workflows. With support for dozens of languages and the ability to handle both text and images, these models operate within Box's secure environment, ensuring enterprise-grade security, governance, and compliance. This development aids global collaboration, helping teams across different regions and languages work more effectively while maintaining Box's trusted security and compliance framework.

Context rot is a phenomenon where providing an AI with too much information leads to less accurate responses, as the AI struggles to prioritize important data within an overloaded context window. This paradoxical effect is particularly pronounced in large language models (LLMs), which have defined context windows that limit the amount of information they can process effectively. Context rot can cause AI to deliver outdated, incorrect, or irrelevant information, posing significant challenges in enterprise settings where accurate data retrieval is crucial. Retrieval Augmented Generation (RAG) is proposed as a solution, offering a method to combat context rot by retrieving only the most relevant information before generating a response. This approach ensures that AI tools access current and pertinent data, reducing the likelihood of errors and enhancing scalability by allowing dynamic updates to knowledge bases. Understanding the mechanisms behind context windows and implementing strategies like RAG can enhance the reliability and accuracy of AI systems, transforming vast amounts of data into actionable insights.

Nomura Research Institute (NRI), a consulting firm with a mission to create a sustainable society through customer collaboration, has leveraged Box's Enterprise Advanced platform to revolutionize its operations with AI-driven workflows. Since adopting Box in 2019, NRI has transformed manual processes into AI-powered operations, significantly enhancing productivity and freeing staff from repetitive tasks like data entry and IT support. The integration of AI has enabled the company to automate compliance-heavy document creation, streamline data entry with automated metadata extraction, and address common IT queries via AI agents, thus allowing employees to focus on strategic tasks. These advancements have not only improved efficiency but have also aligned with NRI's broader goal of creating business value through technology innovation. As pioneers in Japan for implementing Box's advanced AI capabilities, NRI's experience highlights how combining secure cloud platforms with AI can drive organizational transformation beyond simple efficiency improvements.

Creating AI agents that interact with real-world systems can be simplified using Google's Agent Development Kit (ADK), as demonstrated in a tutorial that outlines building a Box agent integrated with Box's Intelligent Content Management platform. The ADK employs a hierarchical structure with tools, sub-agents, and a root agent, allowing developers to create agents that can authenticate with Box, navigate its storage, and utilize its AI capabilities for document analysis. The tutorial emphasizes designing focused tools that handle specific tasks like authentication and data retrieval, which are crucial for establishing connections with external systems like Box. Sub-agents combine these tools with large language model (LLM) intelligence to understand user intent and execute tasks such as file searching and content analysis. The root agent orchestrates the entire process, serving as the main entry point and ensuring seamless interaction with Box's features through clear, structured code.

Gen Z, the most populous generation, is set to inherit substantial wealth, making them a critical segment for the financial services industry to address. This generation, born between 1997 and 2012, has grown up in a digital environment and expects fast, seamless, and personalized experiences akin to those offered by big tech companies. They are frustrated with traditional financial processes, which they perceive as slow and opaque, leading to a potential risk for banks and financial institutions that do not adapt. Gen Z favors real-time customer support across various channels and seeks instant gratification and personalized financial recommendations. To meet these expectations, financial firms need to leverage intelligent content management systems and AI to break down data silos, automate processes, and provide secure, collaborative platforms. This shift not only accelerates processes like loan approvals and wealth management but also ensures compliance with regulatory standards, positioning these firms to better serve Gen Z as they gain increasing financial influence.

Note-taking is a crucial skill for understanding and retaining information, applicable from education to professional settings. Various methods like outlining, Cornell, boxing, charting, mapping, sentence, and Zettelkasten cater to different learning styles and types of content, offering benefits such as clear organization and enhanced recall. The choice between handwritten and digital notes depends on personal preference and learning objectives, with digital tools providing advantages like AI-driven organization and easy editing. Effective note-taking involves not just capturing information but also reviewing, questioning, and engaging with the material to deepen understanding. Technologies and platforms, such as Box, facilitate secure and efficient digital note-taking, offering features to streamline the process and support long-term learning.

Markdown, a lightweight markup language, plays a crucial role in the success of enterprise AI solutions by acting as a universal translator between human-readable content and AI systems. In an episode of the Box AI Explainer Series, Meena Ganesh and Ben Kus discuss how traditional file formats like PDFs and Word documents often contain excessive stylistic data that complicates AI processing, whereas Markdown strips away this clutter to provide clean, structured data that AI can easily interpret. This conversion is essential for enterprises investing in AI, as it ensures that content is accessible and compatible with AI systems, ultimately enhancing the performance and value of AI technologies. The conversation emphasizes the importance of preparing content in AI-friendly formats to avoid costly failures and maximize the effectiveness of AI investments.

Box AI is set to enhance its document creation capabilities through integration with Anthropic's Claude Skills API, allowing users to transform unstructured data into professional documents like PowerPoint presentations and Excel spreadsheets directly within Box. This integration leverages Claude's AI-driven file creation abilities to automate and streamline workflows, reducing the manual effort required in document generation while ensuring consistency and security by maintaining existing permissions. Claude Skills, which comprise organized instructions and resources, enable AI agents to perform specialized tasks, offering a progressive disclosure that loads only the necessary skills for each task, thus optimizing efficiency and speed. This collaboration promises to revolutionize how organizations manage and create content by packaging procedural knowledge and workflows into reusable capabilities, expanding the potential for automation and efficiency in document management across various industries.

CNM, a technical advisory firm for small and mid-sized businesses in the US, embarked on a transformative journey to address inefficiencies in its workflows, particularly in contract management and client onboarding processes. As part of its vision to become a "firm of the future" and enhance trustworthiness, CNM integrated AI capabilities from Box, which led to a 40% increase in client onboarding speed by automating contract processes. Previously, CNM faced challenges like limited visibility into contract statuses and manual follow-ups that caused delays. By using Box's workflow automation tools and encouraging a culture of experimentation, CNM improved transparency and efficiency, reducing guesswork for teams. The firm plans to further integrate automation tools and AI features, aiming for greater operational efficiency and exceptional client service.

Box is enhancing its collaboration with Amazon by integrating its Box MCP server with Amazon Quick Suite, providing a unified AI-driven experience that improves decision-making capabilities. This integration allows enterprises to securely access and manage Box content directly from the Amazon Quick Suite chat interface, maintaining Box's stringent security measures and permissions. It eliminates the need for complex integrations, allowing users to generate, interact with, and save new content all within the Amazon Quick Suite, and ensures that all interactions are validated against Box's permissions. Users can develop custom workflows and AI solutions tailored to their business needs, such as automated project management and legal document analysis, without disrupting existing workflows. The integration supports tools like Anthropic Claude, GitHub Copilot, and Microsoft Copilot Studio, with future support for Salesforce Agentforce, thereby facilitating seamless interaction and AI insights directly from Box content in the Quick Suite environment.

John Allen, Baylor University’s CIO and CISO, advocates for a balanced approach to AI integration in higher education, emphasizing the need for AI to enhance rather than replace human interaction, while simultaneously ensuring robust data security. Under Allen's leadership, Baylor empowers its staff and students to leverage AI tools to improve efficiency and innovation, fostering an environment where technology is accessible and not intimidating. Allen underscores the importance of cybersecurity, stressing that data protection should extend beyond intellectual property to include the identities of students and staff, and advocates for centralized platforms to mitigate risks. He believes that AI can take on routine tasks, allowing human talent to focus on higher-value work, and that this cultural pivot is essential for universities. Allen also highlights the need for continuous vendor assessments to maintain operational integrity as AI tools and regulations evolve. Ultimately, Allen and Box Chief Customer Officer Jon Herstein envision a future where higher education institutions can transform by thoughtfully integrating AI, balancing accessibility, ethics, empowerment, and strategic alignment.

BoxWorks 2025 highlighted Box's strategic direction toward enhancing content intelligence, connectivity, and security, with a focus on features beneficial to law firms. The conference showcased AI-driven capabilities that streamline processes like contract access, insight extraction from case filings, and client onboarding automation, thereby allowing lawyers to concentrate on substantive legal tasks. Box introduced enhanced digital workplace integrations to synchronize legal documents and communications, reducing manual data transfers and version discrepancies. Significant announcements included the launch of Box AI Agents for insightful data organization, Box Automate for workflow automation, and Box Shield Pro for advanced content protection. These innovations facilitate document standardization, quicker client delivery, and robust security and compliance measures, ultimately improving legal workflow efficiency. Legal professionals at BoxWorks discussed AI adoption and explored how Box's capabilities could transform legal operations, emphasizing the importance of modernizing workflows with an AI-powered content platform to enhance productivity and reduce costs.

Box Sales teams have significantly enhanced their sales processes by integrating purpose-built Box AI Agents into their workflow, resulting in a dramatic 90% reduction in RFP response time and a 67% increase in customer engagement. These AI Agents, trained specifically on Box's product information, streamline tasks that previously required extensive manual effort, such as researching accounts, preparing for meetings, and generating personalized outreach plans. By automating routine tasks and providing tailored customer insights, the AI Agents allow sales professionals to focus on building relationships and engaging in strategic conversations. The AI Agents seamlessly integrate with existing Box tools, maintaining compliance and data security, which is crucial for industries like financial services and healthcare. This approach not only improves efficiency but also enhances the quality of customer interactions, showcasing technology as a means to augment rather than replace human expertise.

Box has announced the launch of its AI Agent for Gemini Enterprise, in collaboration with Google Cloud, designed to transform how businesses utilize their enterprise content by integrating intelligent agents into automated workflows. This innovation allows businesses to securely access, search, and synthesize information from their content while maintaining compliance with existing security protocols. The Box AI Agent is built with the Agent2Agent Protocol (A2A) and Box MCP server, enabling interoperability with other agents in Gemini Enterprise, which facilitates the creation of sophisticated, multi-agent systems. It empowers users to extract structured data, generate documents, and store findings back into Box, thus completing the workflow loop and enhancing digital transformation efforts across various sectors such as Financial Services, Life Sciences, and Public Sector. Available through the Google Cloud Marketplace, this agent aims to unlock the potential of unstructured data, ensuring businesses can build smarter, faster workflows by turning content into actionable insights.

AI agents, while democratizing access to expertise, have paradoxically heightened the value of human domain knowledge within organizations, as those with deep expertise extract more value from AI than those lacking it. The success of AI integration hinges on domain experts directing and evaluating AI agents, as they possess the experience to refine prompts, judge output quality, and seamlessly integrate AI into workflows. Companies achieving significant productivity gains do so by upskilling existing experts to combine domain mastery with AI fluency, illustrating that AI amplifies existing skills rather than replacing them. Effective AI use requires a comprehensive understanding of tasks, human oversight, and process optimization, which depend on deep domain knowledge. Research by Box highlights that companies reporting high productivity gains focus on training current employees to bridge AI skills gaps, rather than hiring new talent. This approach emphasizes that AI fluency layered onto domain expertise creates more value than AI fluency alone, with organizations that invest in both areas poised to gain a competitive edge.

The recent update to the Box open-source Model Context Protocol (MCP) server introduces 17 new tools designed to streamline the management of collaborations, which are essential for setting access permissions in Box. These tools are categorized into user lookup, group lookup, and collaboration tools, allowing users to efficiently manage file and folder access permissions by adding or modifying user and group roles such as editor, viewer, and owner. This update aims to simplify the process of managing file permissions programmatically without the need for complex API calls or manual interactions with the Box web interface. By handling authentication, error management, and rate limiting, the MCP server allows users to focus on executing tasks more effectively. This enhancement not only addresses a common developer pain point but also sets the stage for more advanced collaboration workflows, encouraging community contributions to further refine the server's capabilities.

The text appears to be a snippet of code or a script related to a form handling system, possibly for a marketing or customer data platform. It outlines various functions and methods for managing form elements, including setting the order of form rows and columns, validating form inputs, and handling form submissions. The code also includes mechanisms to attach and detach validation and submission callbacks, allowing for dynamic interactions and response handling when forms are submitted. Additionally, there are parts of the code that seem to manage the form's success handling by allowing functions to be executed upon successful form submission.

The recent updates to the Box open-source MCP server address significant issues by introducing three new features: SSE transport, HTTP transport, and CCG authentication, which enhance local development setups by offering more flexible authentication and transport options. CCG authentication replaces the cumbersome browser-based OAuth flow, allowing Box admins to create service accounts with precise permission controls, thus streamlining access without user intervention. Additionally, the updates resolve compatibility problems by incorporating HTTP and SSE transport options, enabling the use of various MCP clients that do not support the original STDIO-only approach. These improvements facilitate the integration of tools like n8n, allowing developers to build workflows that interact with Box content without encountering previous limitations. These updates are practical solutions that eliminate workflow interruptions and enhance the flexibility and usability of the server in local development environments.

OpenAI has introduced AgentKit, a suite of tools aimed at enhancing the development of AI agents, featuring Agent Builder for creating workflows and Evals for testing agent performance. By collaborating with Box, AgentKit enables the creation of secure, content-aware AI agents that operate within an enterprise's proprietary data environment, facilitated by the Box MCP server for content-centric automation. This integration allows sophisticated tasks like data extraction and document analysis while maintaining data security. Evals provides a framework for evaluating and optimizing agent performance, reducing manual effort and ensuring reliability, which is crucial for enterprise adoption. The partnership between OpenAI and Box addresses the challenges of integrating powerful AI with secure data, paving the way for the widespread adoption of agentic AI in business environments.

OCR invoice processing utilizes optical character recognition technology to automate data extraction from invoices, transforming unstructured text from scans, PDFs, and images into structured, machine-readable information that integrates seamlessly with financial systems. This process enhances efficiency, accuracy, and scalability while reducing manual data entry and associated costs. The seven-step workflow of OCR invoice processing includes digitization, image preprocessing, text recognition, data extraction, validation, data export, and archiving, ensuring a secure and compliant audit trail. Despite challenges such as compliance requirements, handwritten content, diverse vendor formats, exceptions, low-quality scans, and integration issues, solutions such as AI-powered workflows and adaptable OCR systems help mitigate these issues. Platforms like Box combine OCR with AI to enhance invoice management by providing scalable cloud storage, content management, and improved vendor relationships, while also offering security and compliance features. OCR technology is generally reliable and accurate, although it may struggle with handwritten invoices and poor scans.

Under the leadership of Suma Nallapati, Denver Technology Services (DTS) is transforming the way city government operates by modernizing its technology infrastructure and embracing innovative solutions like Box for Intelligent Content Management. By adopting Box, DTS addressed challenges with outdated legacy systems, improving record management, data protection, and collaboration across over 50 city agencies. The integration of AI tools, such as the development of a multilingual chatbot named “Sunny,” has enhanced service delivery and earned recognition for innovation. Nallapati emphasizes the role of AI in advancing operations and governance, particularly through better metadata management and automation, which supports flexible, secure collaboration and compliance. The successful implementation of these technologies has led to a significant increase in user adoption, reflecting the productivity gains and enhanced security protocols that benefit Denver's residents.

Leveraging AI in content management enables the transformation of unstructured data into structured, actionable insights, as demonstrated by a workflow that processes insurance policy documents and attaches structured data to a Box document. The process requires Box app OAuth credentials, a Box MCP server instance, and n8n with OpenAI API access, utilizing Box's enhanced extract endpoint for efficient data extraction. Users create metadata templates via the administration console, which are then used as extraction parameters. These templates, defined by simple key pair values, can be associated with files to create metadata instances. The workflow involves locating files using MCP tools, processing each file through a Split node, and determining whether to update existing metadata or create new instances. Box's specialized AI extract endpoint efficiently identifies template-defined key pair values within document content, facilitating the attachment of extracted metadata to files. This approach automates the transformation of insurance documents into structured, queryable data, enhancing downstream automation, compliance monitoring, and business intelligence, while retaining the ability to manage document variations and edge cases.

Financial institutions are increasingly utilizing AI to transform unstructured documents into actionable insights, enhancing client experiences and operational efficiency. By applying AI to high-friction processes such as policy issuance, claims, and onboarding, businesses can automate workflows, improve risk management, and offer faster, more secure services. Tools like Box Shield Pro and Box Extract enable classification and extraction of critical metadata, helping firms maintain audit readiness and enhance threat response. AI-driven solutions, such as those provided by Box, allow institutions to organize and secure data, ensuring that only necessary information is accessible to teams, thereby reducing friction and speeding up processes. The approach emphasizes keeping humans in the loop for judgment-intensive tasks, facilitating a pragmatic, scalable integration of AI into financial services while maintaining flexibility to adapt to evolving technologies. As these capabilities advance, institutions are encouraged to pilot AI applications, learn from initial implementations, and expand their use across more regulated and customer-facing operations.

AI systems, while designed to be helpful, are increasingly vulnerable to manipulation through prompt injection and data poisoning, as discussed by Box's CTO Ben Kus and Senior Product Marketing Manager Meena Ganesh. Prompt injection involves inserting malicious commands into AI interactions, exploiting the systems' eagerness to assist, while data poisoning corrupts AI decision-making by introducing false data, often through emails or documents. As AI agents perform critical tasks, they become attractive targets for sophisticated attacks, necessitating organizations to adopt rigorous security measures akin to those used for critical infrastructure. This includes implementing robust access controls, governance frameworks, and threat detection to prevent and mitigate potential damage. Kus emphasizes the importance of controlling what AI agents are allowed to do to avoid catastrophic consequences, urging businesses to build security into AI deployment from the outset to protect against reputational, financial, and legal risks.

LPL Financial, a leading independent broker-dealer, leverages Box's Intelligent Content Management and AI solutions to enhance advisor productivity while maintaining compliance in a heavily regulated environment. By integrating Box Hubs and Box AI, LPL can efficiently process large volumes of unstructured data, such as call transcripts, to extract sentiment and common issues, improving service responsiveness. This approach not only saves time but also ensures regulatory adherence with human oversight and cross-functional governance. As early AI adopters at LPL share their success stories, adoption grows across the network, supported by targeted training and peer-to-peer learning. Looking forward, LPL aims to further transform workflows by embracing agentic AI, which can autonomously take action with human oversight, ultimately driving innovative solutions and exceptional client experiences.

BoxWorks 2025 showcased how Box is revolutionizing the media and entertainment (M&E) industry by providing tools for creative teams to work more efficiently while maintaining robust security. With the introduction of Box Shield Pro, Box offers AI-powered ransomware detection and advanced security measures to protect valuable content such as film scripts and unreleased music from leaks and breaches. The company also emphasizes the role of AI in enhancing creative processes and operational efficiencies, enabling content management at scale by integrating with over 1,500 tools. Box Automate and AI Agents streamline project workflows, allowing teams to focus on creativity without being bogged down by logistical challenges. By supporting both large studios and smaller production companies, Box aims to facilitate seamless collaboration and innovation, ensuring that creativity can thrive in a fast-paced cultural landscape.

Cinco Design, a creative agency in Portland, Oregon, enhanced its operations by adopting Box's Intelligent Content Management to automate and streamline client intake processes. Previously burdened by a cumbersome system of disconnected software and on-prem servers, Cinco faced challenges in file access, version control, and security, which led to inefficiencies and risks. Box's platform offered robust integrations, customizable permissions, and strong security features, facilitating a seamless migration of over 100 terabytes of data. This transition enabled Cinco to automate NDA workflows, centralize client content, and improve project visibility, thus saving time and reducing legal risks. The improved system also allowed Cinco to retire expensive infrastructure, enhance security with Box Shield, and maintain alignment with tools like Slack and Google Drive. Looking forward, Cinco plans to leverage Box for advanced document search and metadata extraction to further optimize their media file organization.