Tag Archives: levels of assurance

Request for Comment and IPR Review: PCTF Assurance Maturity Model Draft Recommendation V1.0

This review period is officially closed. Thank you.

Notice of Intent: DIACC is collaborating to develop and publish a Pan-Canadian Trust Framework™ (PCTF) Assurance Maturity Model to set a baseline of public and private sector interoperability of identity services and solutions.

To learn more about the Pan-Canadian vision and benefits-for-all value proposition please review the Pan-Canadian Trust Framework Overview.

Document Status: This review document has been approved as a Draft Recommendation V1.0 by the DIACC’s Trust Framework Expert Committee (TFEC) that operates under the DIACC controlling policies.

Summary: It is essential that Participants in a digital ecosystem have a way to evaluate the robustness and trustworthiness of transactions within that ecosystem. In order to do so, Participants must share a common vocabulary that describes the level of confidence they can associate with an Entity or transaction, as well as a common way in which to determine that level of confidence.

In the Pan-Canadian Trust Framework™ (PCTF), a Level of Assurance (LoA) represents the level of confidence an Entity may place in the processes and other conformance criteria defined in any given component of the PCTF.  Levels of Assurance are elemental in creating networks of trust. Levels of Assurance models only work if all Participants in a digital ecosystem are able to interpret them consistently. It is therefore critical that all Participants in an ecosystem agree upon a minimum set of criteria for each Level of Assurance. Only then will a Relying Party in that ecosystem be able to properly evaluate the risks inherent in a relationship or transaction, and the Level of Assurance that can be placed in Participants, Credentials, and those transactions. The components of the PCTF describe the detailed conformance criteria that should be used to evaluate such Levels of Assurance in the context of a given PCTF component. This document provides guidance regarding how to use those criteria in order to properly classify Levels of Assurance.

Invitation: All interested parties are invited to comment.

Period: Opens: June 27, 2021 at 23:59 PT | Closes: July 28, 2021 at 23:59 PT

Document: PCTF Assurance Maturity Model

Intellectual Property Rights: Comments must be received within the 30-day comment period noted above. All comments are subject to the DIACC contributor agreement; by submitting a comment you agree to be bound by the terms and conditions therein. DIACC Members are also subject to the Intellectual Property Rights Policy. Any notice of an intent not to license under either the Contributor Agreement and/or the Intellectual Property Rights Policy with respect to the review documents or any comments must be made at the Contributor’s and/or Member’s earliest opportunity, and in any event, within the 30-day comment period. IPR claims may be sent to review@diacc.ca. Please include “IPR Claim” as the subject.

Process:

  • All comments are subject to the DIACC contributor agreement.
  • Submit comments using the provided DIACC Comment Submission Spreadsheet.
  • Reference the included PDF to include the corresponding line number for each comment submitted.
  • Email completed DIACC Comment Submission Spreadsheet to review@diacc.ca.
  • Questions may be sent to review@diacc.ca.

Value to Canadians: The PCTF Assurance Maturity Model will provide value to all Canadians, businesses, and governments by setting a baseline of business, legal, and technical interoperability. The DIACC’s mandate is to collaboratively develop and deliver resources to help Canadian’s to digitally transact with security, privacy, and convenience. The PCTF is one such resource that represents a collection of industry standards, best practices, and other resources that help to establish interoperability of an ecosystem of identity services and solutions. The DIACC is a not-for-profit coalition of members from the public and private sector who are making a significant and sustained investment in accelerating Canada’s Identity Ecosystem.

Context: The purpose of this Draft Recommendation review is to ensure transparency in the development and diversity of a truly Pan-Canadian, and international, input. In alignment with our Principles for an Identity Ecosystem, processes to respect and enhance privacy are being prioritized through every step of the PCTF development process.

DIACC expects to modify and improve this Draft Recommendation based upon public comments. Comments made during the review will be considered for incorporation into the next draft and DIACC will prepare a Disposition of Comments to provide transparency with regard to how each comment was handled. 

Thank you for your support and participation in this review period.

Help Define and Design the Future of Canada’s Levels of Assurance with DIACC

Online transactions, interactions, and service delivery are no longer aspirational end-states. In the wake of COVID-19, digital has become the default for many Canadians and industries. Making that default work long-term means earning and maintaining trust for people and platforms. Levels of assurance (LOAs) create a clear roadmap for developing that confidence, both for teams offering and people accessing services. Learn more about LOAs in our recent post to understand why they matter to Canadians, where we’re at and where we go from here.

“Making sure we have a common way to evaluate and measure the integrity of that data is central to a common bar of acceptance,” DIACC President Joni Brennan explained. To get a better understanding of how LOAs are a core factor in the future and success of digital platforms, we spoke with Joni and George Watt, Partner, Strategy and Lean Innovation Practice at Becker-Carroll. George has extensive experience leading and developing innovation and security with global enterprises.

Why are Levels of Assurance so important right now? 

George Watt: The pandemic accelerated the pace of digital transformation to breakneck speed. Governments have responded with impressive resolve to deliver new digital services safely. The private sector faced similar challenges with lockdowns and restrictions, new customer needs and opportunities to better serve their customers. For some, these transformations are imperative just to remain viable.

These organizations, both public and private, need to be able to safely transact digitally with their customers, and to interact with one- another. They need to be certain the entities they’re dealing with are who they claim to be with a level of confidence commensurate with risk inherent in each transaction. 

Levels of Assurance are about specific agreements between participants in an ecosystem that enable them to understand the level of confidence they can place in those transactions.

When it’s done well it becomes an unambiguous contract that makes it safe for people to transact digitally. 

Joni Brennan: The LOA conversation is important because it is a piece of the conversation that focuses on the verification of information that would be part of a transaction is authentic. The technical conversation is an important part of the puzzle… but making sure we have good data that can be relied upon and have assurance around that data is ultimately critical. 

For example, if I were a bank, I want to know data is good no matter where it came from. 

LOAs are an important piece of the puzzle for making transactions possible. 

What’s a use case where Canadians interact with LOAs? 

George Watt: When citizens interact with these LOA schemes they aren’t aware it’s happening – but they are aware of its impact. Signing into social media is a low risk example that many people do every day – but those service providers don’t truly know who their customers are. Any of the other services Canadians access with those social media digital credentials would also be considered low level of assurance because nothing is done to ensure the account holder is who they claim to be. 

Online banking is a high risk example. Compromise of that information could be devastating and could lead to bankruptcy or worse. That’s why the Canadian banking industry has a very high LOA standard that helps ensure customers are who they say they are, and helps customers protect themselves. 

The consumer doesn’t see these LOAs in explicit ways but sees their impact implicitly. For example, you don’t need to visit Facebook HQ to get an account – it has a low level of assurance – but you do need to visit your financial institution or go through a more rigorous process to prove who you are to get a bank account. They require a higher level of confidence that they know who you are. 

Joni Brennan: The ability to access city plans and interface on a decision like [changing traffic flow to create active streets] would be a lower risk interaction. Learning that there will be some barriers set up for bicycles would be low risk for other citizens or the municipal government. 
Accessing your pension or tax refund would be an example with higher risk. Something that we haven’t solved yet (but relates to digital ID) is access to digital medical records. Ideally, it will be in a way that the patient can access their own records and make those records transferable or portable between doctors, between patients and across different devices. Access to health records in a ubiquitous way is high risk. It’s something we don’t have today that identity and LOA would help with.

What’s the biggest advantage of a strong LOA scheme? 

Joni Brennan: Part of the ‘why’ behind why we’re talking about this is because we have different LOA schemes in Canada today. Those different schemes have created room for challenges and adoption delays, across different regions and different stakeholders. Different parties might be using the same assurance number – but the way they’re calculating it isn’t consistent, leading to variable outcomes. 

A level of consistency and transparency in terms of assurance – that relates to individual capability in a transaction – makes acceptance and confidence in an interaction clearer and helps the economy as a whole. It’s important for banks, governments, telecommunications providers, and users. Strength in a common acceptance and transparency would help and that’s what we’re missing right now in Canada. 

George Watt: It’s important that we develop a common understanding of these risks and establish a shared vocabulary that ensures we all evaluate risks the same way. Strong LOA schemes will enable positive economic and social impact through more robust delivery of services across domains. 
If we don’t understand those risks collectively, we either won’t deliver the services necessary to live up to our potential — or we will deliver them without understanding the risks, which could be even worse and potentially set us back. Trust is speed. A strong scheme means faster delivery of more robust and trustworthy services. Participation in these digital ecosystems will drive better seamless services for all Canadians.

Why is DIACC advocating for stronger and consistent LOAs?

Joni Brennan: We need commonality in terms of how information is verified. For that measure of assurance of information, we need a common scheme that works across the different schemes that exist in Canada today. It will create visibility and a common approach so that no matter the industry, teams can work from the same starting point for validation and verification of information. That is so important, whether it’s health or AI or smart cities. 

The current scheme and current state of the art doesn’t provide the level of dynamism required in a hyperconnected ecosystem. The current ‘1 through 4’ scheme applied on top of a complicated transaction involving many partners with different capabilities – that singular number is actually insufficient. It’s much better to have transparency, visibility, and a ‘score card’, if you will, that measures assurance that is verified. That’s the kind of dynamism and transparency we need in a hyperconnected ecosystem, that provides scalability in an LOA scheme. 
George Watt: What we had was good – but it needs to evolve to keep up with what we’re dealing with now. We need to solve tomorrow’s challenges today, not yesterday’s problems. A more modern approach to LOAs is necessary to make that happen.

What will be the biggest factor for success? 

George Watt: Bringing the public and private sector together and bridging the many international standards groups… I think the defining factor for success will be collaboration. There are lots of smart people who’ve been thinking about this. More importantly this assurance scheme will work best when private and public sector, NGO and standards orgs work together to create a more trustworthy, more robust ecosystem that allows Canada to live up to its potential. Collaboration will be key. 

Joni Brennan: Collaboration will also represent a diversity of stakeholder needs and values – which is important to ensure the way forward is as inclusive as it can be. Success requires communication and education around the why – why we’re doing this work, the value, as well as how this work will be adopted. 

As George said, for people participating in a transaction these LOAs are meant to be invisible. They’re not always the most exciting or technical part of the work – but they provide that layer of integrity underneath the technology and user experience. To succeed, we’ll need education and communication.

George Watt: Diversity is the rocket fuel of innovation. Working with DIACC, I’ve always been impressed by the diversity of membership and those who participate. It’s a diverse group of smart people who are willing to come together to work on important and complex problems.

Bring your voice to the DIACC and share your perspective on how we can solve these pressing, complex challenges. Together, through our Five Year Strategy, we’re aiming to identify key policy and regulatory enablers and barriers to digital identity growth, including creating a unified approach to LOAs. Join us and subscribe for more on LOAs in Canada.

The Next Evolution of Levels of Assurance in Canada

Levels of Assurance (LOA) play a foundational role in the world of standards, digital identity, and digital transactions. Put simply, LOA is the degree of confidence in the validity of a claim, process, or authentication. In the sphere of digital identity, it is a necessary model to verify that the person or entity claiming an identity is the entity to which that identity was assigned. 

Most Canadians don’t think too deeply about LOAs, and yet most Canadians interact with these models, unknowingly, at some point in their lives. For example, Canadian experience LOAs when opening a bank account, demonstrating qualifications for a government service or benefit, making an insurance claim, or wiring money to a client or family member. 

Organizations that use LOAs to inform their policies and processes often have dedicated strategies and teams, working out contingencies and approaches to maximize security in Canada and internationally. These teams often face challenges interacting with other service providers, meeting different standards across jurisdictions, and minimizing friction for clients accessing their services.

How the Current LOA Model Works

Imagine two people, Samir and Aiya, are trying to apply for a small business loan. Both women have very strong credentials, a passport, driver’s license and the requisite business records. Samir and her credentials are linked through a knowledge-based authentication (KBA), and are accepted after answering a security question she previously populated about her father’s middle name. Aiko and her credential are linked with an in-person ceremony, as she went to her local bank branch with two pieces of identification and her business records to complete the loan application. Aiya’s scenario offers a stronger LOA and Samir’s a weaker LOA. Despite these differences, Canada’s federal LOAs currently dictate that Samir and Aiya both have the same assurance. 

In Canada and many places around the world, it is common for LOA structures to combine a number of factors into a single score. The result is an obscured view of the risk factors and authentication. This lack of granularity into the LOAs of specific capabilities is a challenge present in the construct of LOA models around the world. The deciding factor regarding acceptance of an identity comes down to Relying Parties (parties who rely on the validity of identities) who determine their own risk profiles. 

In this case, the relying party is the bank. The banker helping Samir and Aiya also benefits from a stronger LOA as they sign off on the business loan. In addition to building a stronger relationship with the client, they are able to manage their portfolio with confidence.

There is widespread agreement the current LOA model in Canada is inadequate. While LOAs serve a purpose, they are not transparent and dynamic enough to address the myriad digital solutions and scenarios of today. Internationally, single LOA schemes are no longer state of the art and today’s requirements necessitate separate evaluation for specific capabilities. In the DIACC community, there is consensus that there must be separate schemes for credentials and identity, at a minimum, in order to be useful in the widest possible range of scenarios and contexts. An improved assurance model should be capable of asserting identity and credentials at different levels.

Envisioning a New Risk-based Model for Assurance

A risk-based model offers a more enduring, user- and industry-friendly path forward that enables existing LOA schemes to participate while building for a more dynamic and scalable digital ecosystem. The notion of leveraging a risk-based model is highly applicable as the application of LOAs  today are best determined by performing a threat or risk analysis. The risk-based model must address the likelihood and impact of something happening, and the appropriate mitigation approach. 

LOA is essential in determining liability and risk; offering a clear understanding how a Subject (customer or citizen) and a Relying Party (company or government service) can validate that they are who they say they are. It is a central component in being able to determine whether a transaction should proceed. 

The risk-based starts by assessing risk first and then the approach drives more value for organizations, as they confront the baseline of their current systems and assess risk realistically. It also helps adopters improve their systems through motivation to  reduce or remove risks through various types of mitigation.

DIACC is on a mission to rapidly deliver a modern, risk-based LOA model that is…

  • Risk-based
  • Directive and illustrative
  • Non-prescriptive in execution
  • Evergreen
  • Deterministic in implementation and assessment
  • Congruent with existing state of the art and best practices
  • Inclusive in support of both the private and public sector
  • Supportive of evolving needs on credentials and bindings

The impact of this evolution is far-reaching, and will ensure that the  Pan-Canadian Trust FrameworkTM is strong and resilient over time. This evolution takes a framework-wide approach to address interdependencies, independencies, and support communication across platforms. This new approach ensures scalability over time as technologies and their uses evolve.


DIACC has engaged a small, representative team to rapidly deliver a new model to support the PCTF, which launched September 15, 2020. The model will benefit from the DIACC’s well-documented peer and public review process. Members can contact info@diacc.ca to contribute. Non-members can get in touch to learn more.