We are handing AI agents digital wallets and the autonomy to spend - but we haven't agreed on the rules of the road. While the industry fixates on the capabilities of Agentic Commerce, a massive, unaddressed risk crisis is looming.

Speed is increasing, but so is the attack surface. How do you distinguish a legitimate buying agent from a high-speed fraud bot? If an agent hallucinates and orders 5,000 units instead of 50, who is liable?

Currently, four competing standards are racing to solve this not just as a payments problem, but as a risk and identity problem:

• Visa (TAP) is betting on Identity Verification: verifying the who behind the agent.

• Google (AP2) is betting on Intent Mandates: cryptographically proving the what and why.

• Mastercard (Agent Pay) is betting on Tokenization: restricting the how through scoped tokens.

• Stripe & OpenAI (ACP) are betting on Standardized Discovery: structuring the where to reduce friction (and ambiguity).

In this session, we cut through the hype to ask the hard questions about risk in the agent economy:

1. The New Attack Vectors: How does "agentic" fraud differ from traditional programmatic attacks?

2. The "Trust" Divergence: Why are these four giants taking such different approaches to securing agent transactions? (Identity vs. Tokenization vs. Mandates).

3. The Liability Gap: When an autonomous agent makes a purchase, who owns the risk? The user? The agent developer? The merchant?

4. Strategic Defense: How can platforms build agentic experiences today that don’t become fraud nightmares tomorrow?

The Goal:

Move the conversation beyond "how to build an agent" to "how to survive the agent economy." You will leave understanding which risk framework aligns with your business model and how to prepare for a world where your customer is code.