Compromising privacy by leaking sensitive information

Taxonomy of Risks posed by Language Models

Weidinger et al. (2022)

Source DOI

Sub-category

Risk Domain

2Privacy & Security

2.1Compromise of privacy by leaking or correctly inferring sensitive information

AI systems that memorize and leak sensitive personal data or infer private information about individuals without their consent. Unexpected or unauthorized sharing of data and information can compromise user expectation of privacy, assist identity theft, or cause loss of confidential intellectual property.

"A LM can “remember” and leak private data, if such information is present in training data, causing privacy violations [34]."(p. 217)

Entity— Who or what caused the harm

Human

Due to a decision or action made by humans

AI system

Due to a decision or action made by an AI system

Other

Due to some other reason or is ambiguous

Intent— Whether the harm was intentional or accidental

Intentional

Due to an expected outcome from pursuing a goal

Unintentional

Due to an unexpected outcome from pursuing a goal

Other

Without clearly specifying the intentionality

Timing— Whether the risk is pre- or post-deployment

Pre-deployment

Occurring before the AI is deployed

Post-deployment

Occurring after the AI model has been trained and deployed

Other

Without a clearly specified time of occurrence

Supporting Evidence (2)

"Privacy leaks were observed in GPT-2 without any malicious prompting - specifically, the LM provided personally identifiable information (phone numbers and email addresses) that had been published online and formed part of the web scraped training corpus [34]. The GPT-3 based tool Co-pilot was found to leak functional API keys [109]."(p. 218)

"In the future, LMs may have the capability of triangulating data to infer and reveal other secrets, such as a military strategy or business secret, potentially enabling individuals with access to this information to cause more harm."(p. 218)

Part of Risk area 2: Information Hazards