Last week, CITP’s Tech Policy Clinic, along with Dr. Jennifer King, brought leading interdisciplinary academic researchers together to provide recommendations to the Federal Trade Commission on how it should update the 2013 version of its online digital advertising guidelines (the “Disclosure Guidelines”). This post summarizes the comment’s main takeaways.
We focus on how the FTC should address the growing problem of “dark patterns,” also known as “manipulative designs.” Dark patterns are user interface techniques that benefit an online service by leading consumers into making decisions they might not otherwise make. Some dark patterns deceive consumers, while others exploit cognitive biases or shortcuts to manipulate or coerce them into choices that are not in their best interests. Dark patterns have been an important focus of research at CITP, as noted in two widely cited papers, “Measurement Methods and Dark Patterns at Scale: Findings from a Crawl of 11K Shopping Websites,” and “What Makes a Dark Pattern… Dark?: Design Attributes, Normative Considerations.”
As documented in several research studies, consumers may encounter dark patterns in many online contexts, such as when making choices to consent to the disclosure of personal information or to cookies, when interacting with services and applications like games or content feeds that seek to capture and extend consumer attention and time spent, and in e-commerce, including at multiple points along a purchasing journey. Dark patterns may start with the advertising of a product or service, and can be present across the whole customer path, including sign-up, purchase, and cancellation.
Given this landscape, we argue that FTC should provide guidance that covers the business’s entire interaction with the consumer to ensure that they are allowed to engage in free and informed transactions. Importantly, we highlight why the guidance needs to squarely address the challenge that providing additional disclosures, standing alone, will not cure the harms caused by a number of dark patterns.
Our key recommendations include the following:
- Offer consumers symmetric choices at crucial decision-making points. And offer that parity for crucial decision-making points across different modes of accessing the service so that consumers can exercise the same choices whether they are using web applications, mobile applications, or new forms of augmented reality applications.
- Do not preselect choices that favor the interest of the service provider at the expense of the consumer.
- Disclose material information in a manner that allows consumers to make informed decisions. Consider a heightened requirement to present information in a manner that serves the best interest of the consumer at critical decision points.
- Follow ethical design principles when designing their interfaces. Such principles include taking account how different demographics, especially vulnerable populations, may interact with an interface by testing the usability and comprehensibility of interfaces across the different demographics.
- Disclose if and when businesses use personal data to shape the online choice architecture for users.
Our comments also draw attention to how consumer protection authorities across different countries are addressing dark patterns. While each jurisdiction operates in its unique context, there is a growing set of strategies to counteract the negative effects of dark patterns that are worth drawing lessons from as the FTC develops its own guidance.
We also caution that legalistic, long form disclosures are not read or understood by the average consumer. In other words, relying on boilerplate disclosures alone will not cure most dark patterns. Instead, we point to studies that demonstrate how disclosures shown close to the time of the decision and relevant to that decision are a lot more effective in educating consumers about their choices. As a way forward, we recommend that consumer-centered disclosures should be (1) relevant to the context of transaction, (2) understandable by the respective consumer audience/segment, and (3) actionable, in that the disclosure needs to be associated with the ability to express an informed decision.