CTO Arun Kumar discusses how Socure leverages AWS and generative AI to collect billions of data points each day in order to combat sophisticated online fraud at scale.
Topics Include:
- Introduction of Arun Kumar, CTO of Socure
- What does Socure specialize in?
- KYC and anti-money laundering checks
- Mission: eliminate 100% fraud on the internet
- Fraud has increased since COVID
- Socure blocks fraud at entry point
- Works with top banks and government agencies
- CTO responsibilities include product and engineering
- Focus on increasing efficiency through technology
- Two goals: internal efficiency and combating fraud
- Countering tools like FraudGPT on dark web
- Measuring success through reduced human capital needs
- Fraud investigations reduced from hours to minutes
- Improved success rates in uncovering fraud rings
- Detecting multi-hop connections in fraud networks
- Question: Who's winning - fraudsters or AI?
- It's a constant "cat and mouse game"
- Creating a fraud "red team" similar to cybersecurity
- Partnership details with AWS
- Amazon Bedrock provides multiple LLM options
- Building world's largest identity graph with Neptune
- Real-time suspicious activity detection
- Blocking account takeovers through phone number changes
- Success story: detecting deepfake across 3,000 IDs
- Collecting hundreds of data points per identity
- Challenges: adding selfie checks and liveness detection
- Future strategy: 10x-100x performance improvements
- Creating second and third-order intelligence signals
- Internal efficiency applications of generative AI
- AI-powered sales tools and legal document review
Participants:
See how Amazon Web Services gives you the freedom to migrate, innovate, and scale your software company at https://aws.amazon/isv/