Your Agent Just Authorized What?! — Jay Mok & Ben Coumes, Paypal
PayPal engineers Jay Mok and Ben Coumes present a framework for how AI agents should be authorized to take actions on behalf of users.
The framework centers on three fundamental questions: did the user give consent, is it permitted right now, and can we prove it later? The answers depend on the level of risk involved and whether the parties know each other. The presentation progresses from simple cases (a coding agent with one-time approval) through medium complexity (money transfers between known parties using a shared vault) to the most challenging scenario (agents conducting transactions with completely unknown counterparties). For the most complex case, they propose a technique allowing both the seller and payment processor to verify the transaction without seeing each other's details.
Vibekollen prepared this summary with AI from the original publication. The content belongs to AI Engineer.
More to read
Lumaril, which develops AI software for optimizing industrial production by connecting to existing control systems, raised a $66M Series A at a $400M valuation
Techmeme 3 h ago
DeepSeek open sources Super-Agent as a research prototype, a proactive coding agent it says can monitor group chats for coding issues and suggest solutions
Techmeme 4 h ago
Insurance software maker Guidewire agrees to acquire Hyperexponential, which makes AI-powered insurance underwriting and pricing software, for $1.7B in cash
Techmeme 4 h ago
One word for AI agents in 2027
Nate Herk 8 h ago