Responsible data governance in smart cities: Waterfront Toronto

On 8 December we talked to Anna Artyushina for a practice example interview. Find the recording of our interview and the narrative below.

Waterfront Toronto

In 2017, Waterfront Toronto, a public body created by three levels of government to develop the shoreline of Lake Ontario in Canada, put out a Request for Proposal (RfP) to develop a 1.8 hectare stretch of coastline into a new community—a three-year saga that ended abruptly with unresolved issues around data sharing and privacy. While the primary focus of the original RfP was on developing a viable model of mixed-income housing with a particular focus on moderately priced rental housing, debate and critique of the project focused on what was originally a relatively small component of the original RfP: smart technologies (and the necessary data frameworks to manage those technologies).

A smart city solution by Sidewalk Labs

In response to the RfP, Sidewalk Labs, which is an independent urban planning and infrastructure subsidiary of Alphabet (part of Google) laid out a vision for development that emphasised a new neighbourhood that they described as a “bustling digital and civic workshop open to all” and a “neighbourhood built from the internet up”. Connectivity and data came together to produce a vision of a neighbourhood that would rely on new modes of transportation through self-driving shuttles, thermal grids would provide a grid-wide efficient heating for buildings, and modular housing to reduce building costs. While the original spirit of mixed-income housing and sustainability remained a part of the Sidewalk Labs vision, the primary vehicle for delivering these goals would be data-driven—an idea that cannot be found in the original RfP.

Data-driven frameworks

While Sidewalk Labs vision for a new neighbourhood largely addresses buildings and the connections between them, it also briefly touches on how data could alter regulatory frameworks. In their vision of the project, Sidewalk Labs addressed how data could influence building codes and city ordinances, citing predictive modelling techniques and real-time monitoring that would allow city officials to fine tune requirements around issues such as structural integrity, daylight access, air quality, noise levels, and energy usage. Sidewalk Labs proposed that all buildings in the neighbourhood be equipped with sensors to collect data for analysis.

The image of dynamism

Connected to these data-driven frameworks is an image of dynamism, with Sidewalk Labs looking to replace “20th century static regulations” with “performance-based regulation” and “outcome-based codes” to govern the built environment. In a sense, Sidewalk Labs was arguing that they could bring the dynamics of the digital environment to the street level, including to physical infrastructure to which one does not normally associate a fast-changing pace.

After two and a half years, project folds

After more than two and a half years, Sidewalk Labs pulled out of the project having done no more than conceptualisation and negotiation with various stakeholders. Ostensibly, the official reason that Sidewalk Labs walked away from the project was over economic uncertainties created by the start of the COVID-19 pandemic; however, the project faced a few obstacles, the most significant being distrust from stakeholders on the intentions of Alphabet and Sidewalk Labs in terms of data. In 2019, Waterfront Toronto conducted a consultation process, which included a survey of approximately 1,000 Torontonians as well as a number of public meetings, and they found concerns around “data collection, surveillance, and inability to get informed consent from citizens”. More specifically, some stakeholders expressed concerns about “boutique deals” on data governance with distrust over the “true objectives/ambitions” of Alphabet and Google.

Data trust fail to produce trust

Given the levels of distrust right to the end of the project, it might leave the impression that the project failed to address how to increase the trust of potential residents in an environment with large levels of data collection. However, Sidewalk Labs’ did develop a concrete proposal for data governance in 2019 with its proposed “Urban Data Trust”, a governance model to approve and manage data-collection devices while also governing algorithmic decision-making. The trust would be a non-profit, independent body represented by a five-person board, including a chief data officer that would be responsible for developing a charter and Responsible Data Use guidelines while also developing oversight and review processes.

A problematic governance model

According to researchers Lisa Austin and David Lie from the University of Toronto, this governance model was problematic because it pursued two distinct strategies to protect data privacy. On the one hand, they describe the fact that the data trust relied on an open-data model with data being largely publicly available and anonymised “by design”. On the other hand, the governance model acknowledged that personal data would be collected and would need to be anonymised in some way to protect individual privacy. The inherent contradiction that these researchers saw was that an open-data model implies no ownership over data and, as such, data would be collected in such a way that it would be anonymised by design. But, if the governance model acknowledged the need to govern algorithmic decision-making, this is already an acknowledgment of the risks to individual rights and the fact that data protection laws—premised on the idea of individual control over data—would still apply.

The shadow of the rentier economy

Researcher Anna Artyushina takes this argument one step further, arguing that the Sidewalk Labs proposal aimed to reclassify data collected from individuals into a common good (which would include de-identified data, aggregated data, and non-personal data). They created a definition of “urban data”, which they classified as information “gathered in the city’s physical environment, including the public realm, publicly accessible spaces, and even some private buildings”. Importantly, this data was classified as “transactional”, meaning it would not be governed by the trust and be owned by the data collector. The concern from researchers was that Sidewalk Labs was looking to at least partially circumvent data privacy laws and to monopolise the ability to capitalise and monetise data collected in the neighbourhood.

The role of public-private partnerships

The role of data governance and building public trust provides lessons for other smart city projects, where data integration provides opportunities for innovative solutions to urban problems but also presents concerns for those concerned about data privacy and potential discrimination. As public and private actors work towards models that balance profitability with social concerns and fundamental rights, it also raises questions about the capacity building that needs to take place within public bodies so that they can best protect public interests and create realistic proposals for urban development that balance various interests.


The City Institute of York


No sector focus






2021 - ongoing


Business model

Academic research


Anna Artyushina

Type of organisation


Data sharing model(s)

Stakeholders involved in smart city projects.

Core impact

York University studies the role of data governance in building public trust for smart city projects, focusing on opportunities for innovative solutions, data privacy, and potential discrimination.


Anna’s dissertation research and the forthcoming book explore a range of policy, civic, and technological initiatives, designed to facilitate responsible data governance in the smart cities in North America and Europe. She analyses the changing role of the state in a digital economy and the governance challenges brought on by the adoption of biometric technologies, AI, and data trusts.

For questions and comments, please visit our forum on Futurium.