Transformative Outcomes of AI: From Aspirational Goals to Exceeding Expectations
American International Group (AIG) has made remarkable strides in artificial intelligence over the past year, surpassing earlier expectations regarding its capabilities and outcomes.
During an Investor Day event less than a year ago, AIG showcased the potential benefits of generative AI. At that time, CEO Peter Zaffino described these aspirations as “aspirational.” However, in a recent conference call discussing fourth-quarter earnings, Zaffino noted, “We see the capabilities are much greater.”
Related: AIG Underwriting Income Up 48% in Q4 on North America Commercial
“We’re witnessing a massive shift in our ability to process a significant submission flow far beyond our expectations, all without additional human capital resources,” Zaffino remarked. “That has been the biggest surprise.”
Zaffino highlighted that AIG has “made significant progress embedding generative AI across our core underwriting and claims processes, and expanding it,” with the goal of deploying its AI tool, AIG Assist, in the majority of its commercial lines businesses.
For instance, AIG Assist has already been utilized in its excess & surplus company, Lexington Insurance. At the Investor Day, AIG set an ambitious target of reaching 500,000 submissions by 2030, and Zaffino reported that Lexington has already surpassed 370,000 submissions.
“I believe the acceleration and opportunity are greater than I anticipated at Investor Day,” Zaffino stated during the call. He discussed the necessary training to manage incoming data more efficiently and the “orchestration of [AI] agents” to analyze this information without bias throughout the entire workflow.
The implementation of generative AI can significantly streamline the front-to-back workflow, according to Zaffino. He explained that orchestration, in AI terms, refers to the coordination and management of AI models and systems. While AIG did not delve deeply into AI orchestration during Investor Day, the company has developed a framework to coordinate AI agents, enhancing decision-making and reducing costs across the organization.
Zaffino described AI agents as “companions that operate alongside our teams.” These agents can provide real-time information, offer advice based on historical use cases, and even challenge underwriters’ decisions. This orchestration can effectively “streamline simple, repetitive, and lengthy processes to support decision-making,” he added.
Related: AIG’s Zaffino to Step Down as CEO as Aon’s Andersen Steps In
AIG has also leveraged its generative AI capabilities in the conversion of Everest’s retail commercial business. Zaffino noted that accounts were prioritized for renewals “in a fraction of the time.” The insurer developed an ontology of Everest’s portfolio, which, when combined with AIG’s own ontology, allowed for effective prioritization of how the portfolios could integrate.
The recent launch of Lloyd’s Syndicate 2479, in partnership with Amwins and Blackstone, marked AIG’s first deployment of generative AI for a special purpose vehicle (SPV). In collaboration with Palantir, large language models were utilized to assess whether Amwins’ program portfolio aligned with the risk appetite of Syndicate 2479. Zaffino expressed confidence in AIG’s “strong pipeline of SPV opportunities.”
Interested in AI?
Get automatic alerts for this topic.
American International Group (AIG) has made remarkable strides in artificial intelligence over the past year, surpassing earlier expectations regarding its capabilities and outcomes.
During an Investor Day event less than a year ago, AIG showcased the potential benefits of generative AI. At that time, CEO Peter Zaffino described these aspirations as “aspirational.” However, in a recent conference call discussing fourth-quarter earnings, Zaffino noted, “We see the capabilities are much greater.”
Related: AIG Underwriting Income Up 48% in Q4 on North America Commercial
“We’re witnessing a massive shift in our ability to process a significant submission flow far beyond our expectations, all without additional human capital resources,” Zaffino remarked. “That has been the biggest surprise.”
Zaffino highlighted that AIG has “made significant progress embedding generative AI across our core underwriting and claims processes, and expanding it,” with the goal of deploying its AI tool, AIG Assist, in the majority of its commercial lines businesses.
For instance, AIG Assist has already been utilized in its excess & surplus company, Lexington Insurance. At the Investor Day, AIG set an ambitious target of reaching 500,000 submissions by 2030, and Zaffino reported that Lexington has already surpassed 370,000 submissions.
“I believe the acceleration and opportunity are greater than I anticipated at Investor Day,” Zaffino stated during the call. He discussed the necessary training to manage incoming data more efficiently and the “orchestration of [AI] agents” to analyze this information without bias throughout the entire workflow.
The implementation of generative AI can significantly streamline the front-to-back workflow, according to Zaffino. He explained that orchestration, in AI terms, refers to the coordination and management of AI models and systems. While AIG did not delve deeply into AI orchestration during Investor Day, the company has developed a framework to coordinate AI agents, enhancing decision-making and reducing costs across the organization.
Zaffino described AI agents as “companions that operate alongside our teams.” These agents can provide real-time information, offer advice based on historical use cases, and even challenge underwriters’ decisions. This orchestration can effectively “streamline simple, repetitive, and lengthy processes to support decision-making,” he added.
Related: AIG’s Zaffino to Step Down as CEO as Aon’s Andersen Steps In
AIG has also leveraged its generative AI capabilities in the conversion of Everest’s retail commercial business. Zaffino noted that accounts were prioritized for renewals “in a fraction of the time.” The insurer developed an ontology of Everest’s portfolio, which, when combined with AIG’s own ontology, allowed for effective prioritization of how the portfolios could integrate.
The recent launch of Lloyd’s Syndicate 2479, in partnership with Amwins and Blackstone, marked AIG’s first deployment of generative AI for a special purpose vehicle (SPV). In collaboration with Palantir, large language models were utilized to assess whether Amwins’ program portfolio aligned with the risk appetite of Syndicate 2479. Zaffino expressed confidence in AIG’s “strong pipeline of SPV opportunities.”
Interested in AI?
Get automatic alerts for this topic.
