When Smart Devices Talk: AI and Fourth Amendment Rights

Artificial intelligence (AI) is woven into the fabric of our everyday lives, whether it is smartphones in our pockets or virtual assistants in our homes or offices.

While smart devices aim to make our lives easier, their proliferation raises important questions about privacy.

Today there are a wide variety of smart home products available on the market from top technology brands. Simply saying “Ok, Google” or calling out to Amazon’s Alexa wakes up the devices from the respective companies. Virtual assistants make themselves useful to us by listening, recording, speaking and collecting a wide variety of information about how we live and work. They adopt predictive capabilities and come to recognize whether our behaviors are consistent with established patterns across various aspects of our lives, such as the route we take from work to home or the types of music we listen to.

Like texts and emails, smart devices record and store large quantities of information that is now increasingly being sought after by law enforcement. For instance, archived voice recordings can potentially reveal who arrived at or left a crime scene. Other data can indicate where and when lights were switched on, and whether certain patterns of behavior that occurred were normal or not.

AI in the form of virtual assistants can actually assume the role of a witness, providing crucial evidence in an investigation or lawsuit.

However, the use of evidence obtained from AI raises key concerns about our constitutional rights under the Fourth Amendment.

Constitutional law supports a legitimate expectation of privacy even with our smartphones, which are known to contain vast quantities of personal information. There is no legal basis to view data stored on other smart devices any differently. Most people have a reasonable expectation of privacy when it comes to the smart devices they use in their homes. But does willingly installing a device that is listening to your voice and transmitting information to a third-party cloud automatically mean you have waived your privacy rights under the Fourth Amendment?

While the Fourth Amendment provides protection against unreasonable searches and seizures, it does not stop our home-based smart devices from presenting evidence against us. In some cases, they may even supply investigators with helpful details that end up corroborating an alibi, for instance.

There are various laws that dictate when people are required to submit to third parties the data stored on their computers, tablets, smartphones and other devices. Although the Fourth Amendment does not prevent access to this information, it requires law enforcement to comply with rules specific to context. In the criminal context, law enforcement agencies must obtain search warrants if they want to access recordings or data that are stored on various smart devices. In the civil context, they must demonstrate a high level of need and relevance to a particular case.

There have already been multiple cases where the interplay between AI and the Fourth Amendment have been thrust into the spotlight. In its landmark June 2018 ruling in Carpenter v. United States, the Supreme Court found that police needed to have a warrant in order to obtain cell phone records that detailed the location information generated by the plaintiff’s cell phone. The plaintiff, Carpenter, had sought to suppress the records under the Fourth Amendment. The court found that accessing the information without a warrant was indeed a violation of the Fourth Amendment.

In a paper titled “Artificial Intelligence and Policing: Hints in the Carpenter Decision,” University of California-Davis law professor Elizabeth E. Joh wrote that while AI can play a beneficial role in investigations, its use poses the risk of government infringement on personal rights. “Rapid advances in technology are changing Fourth Amendment boundaries,” she wrote, noting that law enforcement agencies are now able to access substantial amounts of data automatically recorded by AI and stored in the cloud.

In a recent murder investigation, Arkansas police had a search warrant that sought audio recordings made by Amazon’s Alexa at the crime scene. Amazon tried to suppress the search warrant, arguing that the company sought “to protect the privacy rights of its customers when the government is seeking their data from Amazon.” The case was eventually dropped by the prosecutor, but it may be cited during future investigations that involve AI and privacy issues.

Such cases are contributing to establishing important legal precedent, especially in light of the growing capabilities of the smart devices of the future. For now, the intersection of AI and the Fourth Amendment raises important questions about how to strike a balance between public interests and personal privacy, as well as regarding the accuracy of information that smart devices store. As AI develops and further integrates into our lives, these issues will be thrust even further into the forefront.