By Elizabeth Lynch
Meet Marisa Boston. The daughter of Italian immigrants, she attained the highest level of education possible, and as a rising star on KPMG’s Audit Technology team, she's disrupting two industries in which women have been historically underrepresented. The Denver mother of two recently was named to the 2020 IBM Women Leaders in AI list of honorees, celebrating global women leaders shaping the future of artificial intelligence.
Boston is a trailblazer in her field and an example of the future for KPMG Audit Technology. As a lead AI and automation architect, she led a team that filed three AI patents for KPMG, and was co-inventor of a blockchain-based method to store, track and trace data used throughout the AI lifecycle, leading to KPMG’s first blockchain-related patent.
She is helping change the face of the accounting and technology industries. Women held only 26 percent of professional computing positions in 2019, according to The National Center for Women & Information Technology. At CPA firms, women accounted for 23 percent of the partners in 2019, according to the AICPA Trends survey.
Boston earned her doctorate in computational linguistics and cognitive science from Cornell University just as IBM Watson bested Jeopardy greats Brad Rutter and Ken Jennings, opening up possibilities for the field of AI. Boston’s doctoral work focused on developing computer models of how human brains process sentences, and her first position was in a research lab where she worked on training IBM Watson to understand medical question-answering sentences.
Now, in her new role with Audit, she is developing technology that uses artificial intelligence beyond automation to augment human decision-making. She’s using her knowledge of natural language processing to create search engines to help accountants understand risk inconsistences. Her work involves document investigation, which requires specialists to review disparate financial and legal documents to determine risk across a portfolio. These technologies are used to aid users with current challenges in finding information, but are built within workflows that allow the systems to quickly react to changes in content and processes.
For example, extracting data from financial tables may seem like an easy task. But people make many inferences and assumptions when reading tables and those assumptions must be programmed into the technology for AI to read the information properly. KPMG data scientists developed a patent-pending technology that will be used to accurately bring financial table extraction capabilities into the firm’s automated vouching tools.
Having a perspective that accounts for how AI technology interacts with people propels the firm toward more consistent content, curated by the most experienced professionals, while achieving digital transformation efficiencies along the way.
“I love my work,” said Boston. “We get to re-imagine the way our auditors work in a digitally transformed world, while remaining committed to regulations around audit quality and increasing trust in capital markets. This will not be possible without well thought-out and sophisticated technologies, deployed by strong leaders committed to these end visions.”
Working in a heavily regulated industry that requires sophisticated and specialized analysis means finding niches and integrating AI and other technologies at critical steps where technologies can be used to move forward on the spectrum of digital transformation.
“Applying AI principles to today’s digital world is incredibly rewarding and satisfying,” said Boston “It’s important for me as a technologist to understand the boundary between true and applied AI. The “correct" technical solution provided at the wrong time is not always the right solution, but with the right vision and guidance, we can get there.”
For more information or to arrange an interview, please contact Elizabeth Lynch.