What’s Next in Intelligent Document Processing? Unlocking the Power of Your Content

The rapid advancement of artificial intelligence/machine learning (AI/ML) in recent years has captured global attention, particularly with the introduction of Generative AI and OpenAI's latest Large Language Models (LLMs). The ChatGPT website from OpenAI has enabled regular users to engage with these models. This has created fierce competition among US tech behemoths (Microsoft, Amazon, Google, IBM, and others) and startups, constantly releasing new capabilities through updated models.

While the regular release of LLMs is new and exciting, it presents crucial questions for numerous institutions, especially Federal Agencies such as the United States Citizenship and Immigration Services (USCIS). They question if AI/ML breakthroughs, such as LLMs, might help them address pressing issues, like minimizing visa application backlogs, and if so, how? Additionally, there is a concern about using these technologies in a way that complies with the Agency's security and privacy standards.

Moreover, as the next generation of AI/ML capabilities emerges, organizations must consider how this progress will impact their existing solutions. New research and developments in smaller language models show the potential to outperform LLMs. Therefore, anticipate how these advancements might affect the solutions already developed using current technology.

Instabase is actively developing an AI-based platform called AI Hub, which operates on a SaaS model. The primary goal of this platform is to address various challenges, including providing support to federal agencies as part of their roadmap. AI Hub will serve as a repository for AI applications, specifically focusing on content understanding and employing generative AI tools that leverage recent advancements in AI/ML technologies, including LLMs.

The key components of AI Hub are as follows:

1. AI Hub Converse: This feature facilitates interactive conversations with content, enabling seamless communication with the AI platform.
2. AI Hub Build: This allows users to create repeatable end-to-end workflows, streamlining processes efficiently.
3. AI Hub Apps: A set of pre-built applications designed to automate common processes (e.g., identity verification) and handle documents (e.g., passports) efficiently.

Initially, Instabase's AI Hub will utilize OpenAI's LLMs. However, in the future, it will also support other LLMs based on customer requirements. Instabase has integrated with OpenAI in a way that prioritizes customer data privacy. Specifically, OpenAI will not retain any customer data, and no one at OpenAI can review the data sent by customers. Furthermore, customer data sent to OpenAI cannot be utilized to train or enhance the LLM models.

It's worth noting that AI Hub offers additional capabilities not currently supported by LLMs. These include:

●      Encoding layout and style to handle complex use cases like tables

●      Support for long documents and multi-document use cases

●      Support for multimodal content (e.g, signatures, custom visual objects, barcodes)

●      Advanced refinements (e.g., checking against internal databases) and validations

●      Human-in-the-loop review / correction / verification

●      Enterprise Security and Compliance measures

AI Hub's enterprise capabilities for the private sector are anticipated to be released in the last quarter of 2023. Meanwhile, the roadmap includes AI Hub for the Federal Government, and a beta version is already publicly available at aihub.instabase.com. This platform has the potential to significantly impact operations at USCIS and other Federal Agencies, helping them address their challenges more effectively.

For instance, using AI Hub, Instabase can develop a model for identifying and extracting required information from commonly used document types with a single sample, such as the I-129 or I-589 forms. This streamlines the process and saves considerable time and development resources. Once AI Hub for Federal is accessible, Instabase can extend this capability to identify and extract information from any document type, ranging from structured to semi-structured and fully unstructured documents, still requiring only a single sample for development. This opens up the opportunity to process various document types at scale, regardless of whether they are typed or handwritten. With advanced validation capabilities, Instabase can efficiently flag only applications that do not meet USCIS validation rules for human review. The data from the remaining forms can be extracted and seamlessly sent to USCIS case management systems. In summary, AI Hub allows agencies like USCIS to harness AI/ML innovations to address critical challenges at the agency level.

Considering Federal Agencies' needs, it's important to note that commercial entities are eagerly awaiting the release of AI Hub Enterprise to tackle complex document processing challenges. However, specific concerns need addressing for large-scale applications within the Federal Government. OpenAI and Azure OpenAI have not yet secured FedRAMP certification, a requirement for certain Federal Government use cases. As a result, the current SaaS AI Hub offering cannot leverage OpenAI for such cases. Instabase is actively exploring alternative deployment options to overcome this limitation. Additionally, efforts are underway to support Agencies that have deployed LLMs on-premises for their specific use cases.

‍