Data PrivacyUsers’ Privacy Perception and Acceptance of Voice Based Digital Assistants

March 30, 20220


Over the past few years, services backed by Artificial Intelligence (AI) have emerged as an important trend in the digital market. And most importantly, there has been a massive growth in services using Virtual Assistants or Voice Based Digital Assistants (VBDA).

VBDA are software applications with conversational interface that can interpret human voice and based on the instructions provided, perform tasks and respond to the queries. VBDAs such as Alexa, Siri, Google-Assistant, etc. provide users the comfort of getting some of their tasks done by just giving voice instructions.

Initially the VBDAs were intended to be the voice based interactive feature that would allow the users to operate their smart phones, laptops and tabs but now they are being integrated into a vast variety of products and services. Like there are smart homes which are equipped with several interconnected smart devices such, all of which can be monitored by voice instructions.

The users can interact with virtual assistants in a conversational manner or get their tasks done like setting alarms, making calls, playing music, etc.

VBDA’s popularity and acceptability has grown to the extent that there are over 4 billion VBDAs being used in devices all over the world and it’s estimated to reach 8.5 billion (more than world’s population) units by 2024. However, there are several privacy and security issues involved with the information that the VBDA companies store in the clouds.

Characteristics of VBDA

  1. Human personification in its interaction
  2. Natural and Conversational style of Interface
  3. Highly sensitive always-on microphones for always listening
  4. Personalized and efficient service delivery
  5. Hands free, controlled by voice

There are several technologies involved in this process as with any other computing process. The device with which user interacts has a smart speaker which has internal storage and supports WiFi and Bluetooth. During this process whatever interactions take place between the user and service provider are stored on the cloud.


VBDAs collect a lot of data including the personal information of the user, in order to deliver the helpful results. Over the past few years, the concept of virtual assistants has become more common and as a result, the amount of data associated and stored by virtual assistants has proportionately increased.

As the proliferation of VBDA increases in our daily life, the potential risk of them collecting personal information also increases. There have been many instances where virtual assistants have collected the personal information and shared it with other individuals.

One similar incident took place in 2018 wherein an Alexa user from Portland, Oregon, asked Amazon to investigate a matter where her device recorded a private conversation between her and her husband on the subject of hardwood floors and sent the audio to a contact without her knowledge. It was only when the person contacted her; she came to know that her conversation is being shared.

There are various privacy and security concerns around the information that the VBDA companies store on their servers.

The user’s privacy perception towards VBDAs can be divided into three parts:

  1. Perceived Privacy Risk
  2. Perceived Privacy Concern
  3. Perceived Trust

Perceived Privacy Risk

Risk can be defined as the probability of losing something, similarly privacy risk can be defined as a probable loss of personal information that the user might incur from the disclosure of personal information. Though there might be some other concerns associated with privacy, the loss of control over personal information is the major aspect.

There are a lot of debates and discussions going on about how the tech giants like Google, Facebook and Apple collect user’s information and use this information for the purpose of developing AI algorithm. There have been multiple instances where a user’s personal information and privacy has been compromised with which makes consumers more concerned about how virtual assistants collect and use their information without their prior knowledge.

This potential privacy risk results in the users to losing their trust in the service provider. All such instances of breach in privacy affect the user’s perception towards VBDAs. Perceived privacy risk has negative impact on the user’s perception towards the VBDAs.

Perceived Privacy Concern

Privacy Concern is the assessment of a probable loss which the user might incur due to compromised personal information such as misusing or unforeseeable usage of the personal information. Privacy risk and privacy concern are almost similar just that risk being probability of loss/misuse of personal information and concern being the assessment of a probable loss/misuse of personal information.

The increasing sophistication of the technology used in terms of data collection has increased the privacy concerns of the users enormously. Users are mainly concerned about the unauthorised collection and usage of personal data. Specifically talking about the VBDAs, the users might be concerned particularly about the loss of personal information.

For instance, Alexa starts recording as soon as it listens the active word ‘Alexa’, there are chances that you are having some conversation and speech related to Alexa comes up which might activate it and some personal discussion might be recorded, as happened in one of the cases discussed above. Cases like these make users more concerned about their privacy which negatively impacts the user’s perception towards the VBDAs.

Perceived Trust

User’s trust in a company can be defined as the belief regarding the integrity and ability of the company to safeguard their interests. Trust plays a vital role in users’ perception towards adoption of some technology or relying on some service provider.

Trust is an essential element for any kind of transaction especially in the case of privacy concerns related to the VBDAs. Although trust being the essential element of consideration before adoption of new technology; the tech giants have been under scrutiny for violating consumer trust.

Such instances create a doubt on the trustworthiness of the tech giants and this in turn affects the users’ perception towards adoption of voice based digital assistants. Users’ intention to adapt a new technology is directly proportional to users’ trust in technology and service providers.

Although privacy risk, privacy concern and trust play a significant role in the users’ perception and willingness to adopt a new technology but data shows that though users reported high concerns about privacy, they readily provided their personal information for the comfort and style of living, the VBDAs provide.


To provide conversational services to the users, VBDA needs access to large amount of personal information such as location, appointments, schedule, etc. This poses high privacy risk as the VBDAs track and process personal information without users’ active knowledge just after taking initial consent of the user and the algorithms used are incomprehensible to users.

Different service providers have different privacy policies and their privacy control features allows the users to control the information they want to share with the VBDAs. Although there are privacy policies, it will not be fair to say that users have privacy data control while using VBDAs because of the need to provide very sensitive information for the VBDA to work. Studies have shown that in practical aspect users do not have many options available to control their information while using virtual assistants.

Therefore, it becomes important for the users to thoroughly go through privacy policies and choose the VBDA that offers explanation of its decisions regarding information collected and keep users aware about the possible impacts of the data disclosure.


A common feature of all the VBDAs is their reliance on personal data of the users for the purpose of actively performing the tasks. However, the chances of misuse of such collected data by the service providers are very high and can result in data breaches.

In the light of the foregoing, it is of pivotal importance to create awareness amongst the users. Majority of the users of VBDA are unaware or ignorant towards the privacy concerns that come along with the usage of such applications or devices, especially in the absence of the Data Protection Laws.

VBDA service providers and developers of such applications should ensure the incorporation of data privacy measures in order to safeguard the personal and sensitive data of the users.

Team AMLEGALS assisted by Mr. Upendra Yadav (Intern)

For any queries or feedback, please feel free to get in touch with or

Leave a Reply

Your email address will not be published. Required fields are marked *

Current day month ye@r *

© 2020-21 AMLEGALS Law Firm in Ahmedabad, Mumbai, Kolkata, New Delhi, Bengaluru for IBC, GST, Arbitration, Contract, Due Diligence, Corporate Laws, IPR, White Collar Crime, Litigation & Startup Advisory, Legal Advisory.


Disclaimer & Confirmation As per the rules of the Bar Council of India, law firms are not permitted to solicit work and advertise. By clicking on the “I AGREE” button below, user acknowledges the following:
    • there has been no advertisements, personal communication, solicitation, invitation or inducement of any sort whatsoever from us or any of our members to solicit any work through this website;
    • user wishes to gain more information about AMLEGALS and its attorneys for his/her own information and use;
  • the information about us is provided to the user on his/her specific request and any information obtained or materials downloaded from this website is completely at their own volition and any transmission, receipt or use of this site does not create any lawyer-client relationship; and that
  • We are not responsible for any reliance that a user places on such information and shall not be liable for any loss or damage caused due to any inaccuracy in or exclusion of any information, or its interpretation thereof.
However, the user is advised to confirm the veracity of the same from independent and expert sources.