Cardless has launched Cardless AI, adding conversational servicing, virtual cards for AI agents, and a fraud detection model to its card platform.
The three capabilities are available to Cardless partners from launch, allowing them to offer the features within their own apps, websites, and ecosystems. According to Cardless, consumers increasingly use AI agents to research products and delegate tasks, but still have to complete payments themselves, while generative AI is also making it easier for fraudsters to create fake identities and forge documents.
Conversational servicing and agent spending
With conversational servicing, cardholders can ask their credit card questions and give instructions in plain language, such as checking their balance, making a payment, freezing the card, setting limits on merchants or categories, or asking how much they spent on flights in a given year. Cardless partners can add the feature to an existing in-app conversational experience or build one using Cardless’ APIs.
Cardholders can also give an AI agent its own virtual card, with spending limits they set. For example, a cardholder could instruct an agent to buy a specific item once its price falls below a set amount, and the agent could spend up to that limit and no more. Cardless says its virtual cards are ready for agents to use now, and it is building tools that let agents make payments and purchases on the account.
Parallax for identity fraud
Parallax is a proprietary AI model designed to identify AI-generated and altered images, detect synthetic and manipulated identities, and adapt to new fraud techniques. Because most identity verification providers only see the applications they process, fraudsters can spread applications across several providers. Parallax connects signals across providers to link seemingly unrelated applications and identify fraud rings that individual providers might miss.
According to Cardless, about one-third of the fraud rings Parallax identified during beta testing had activity across multiple identity verification providers. The company says the model achieves these results without exposing applicants’ personal data, allowing Cardless and its partners to stop fraud while protecting the privacy of legitimate cardholders.
