Federal Deposit Insurance Corporation

Examiners must analyze a large amount of data to identify regulatory areas with relatively higher risk of violations on a data driven objective basis.

AI can help the FDIC identify instances of misrepresentation of FDIC Deposit Insurance.

Increase in self-service available answers and a reduction in issue resolution times through improved quantity and quality of information available to provide answers to previously solved problems.

Apply a risk model to identified risks and incorporate additional data sets to identify any correlations or possible solutions.

Analyze investment history and provide an additional predictive analysis for investments.

The current trust document review process is labor intensive and may require further reviews by legal experts, hence, prolonging the processing time.

Reduce the time it takes to review the writing assessment and cost of essay scoring for the purpose of making more timely employment decisions.

Reduce the time it takes to find legal research information, so approved FDIC employees can answer the legal sensitive questions faster and easier.

Take in budget data for reporting and prediction.

Summarize divisional efforts into a narrative taking into consideration staff numbers and competencies.

Monitor retention rate, retirement planning, and knowledge retention and succession.

Human capital resources are used on reading and summarizing content, research, generating draft documents, generating briefings, and searching for agency information.

The pre-exam planning process is time intensive and FDIC has limited examiner resources to review pre exam planning materials and make examination scoping decisions

Scoping fair lending examinations can be time intensive.

Currently, claims specialists spend countless hours manually reviewing trust agreements to extract grantor/settlor and beneficiary information.

Searching for information using AI rather than manual reviews of pdf documents or using the RADD search function.

This machine learning (ML) model is intended to identify banks most likely to be downgraded at or before the next examination.

The current process to extract structured data from bank provided PDF documents of consumer loan applications takes FDIC personnel a significant amount of time to manually transcribe data from PDF loan applications, l...

Compiling data from internal FDIC website and targeted documents to answer questions about CFOO admin.

Compiling data from internal FDIC website and targeted documents to answer questions about FDIC travel.