Getting the Most Out of an AI DeploymentSWIFT's Guy Sheppard on the Challenges of Using AI to Enhance Security
Many companies claim to be successfully using artificial intelligence for security, but the use cases are still not convincing because the technology is incapable of detecting unknown malware, says Guy Sheppard, head of the APAC Financial Crime Compliance Initiative at SWIFT, a global provider of secure financial messaging services (see: Using Artificial Intelligence to Fight Money Laundering).
“As far as artificial intelligence is concerned, many companies are claiming incredible use cases, but the challenge with artificial intelligence programs is that they are very dependent on data. A small error in data can vary the results to a great extent," Sheppard says.
In a video interview with Information Security Media Group, Sheppard also discusses:
- The challenges involved in implementing AI and machine learning;
- How to redefine KPIs for AI and ML to achieve efficiency and effectiveness;
- How to best leverage AI capabilities.
Sheppard leads the Asia Pacific financial crime compliance, intelligence and cyber teams at SWIFT. He has more than 16 years of experience in financial crime compliance, mainly in utilities, entity resolution and correspondent banking KYC. Previously at SWIFT, he was responsible for financial crime compliance across the Americas, the U.K. and Nordic regions.