Public content provenance for organisations
Pages
Page 5 of 6
4. Deploying public content provenance systems: considerations and example use cases
4.1 Points for organisations to consider
When considering deployment of a public content provenance system, there are a number of questions organisations should ask themselves.
4.1.1 Strategy to establish public information trust
Public information trust strategies will vary depending on factors such as the subject domain, the audience, and the objectives of actors seeking to use the organisation’s public information against them.
Many organisations already have some capability for establishing trust in their public information and countering claims made against them. Using public provenance will help establish trust for an organisation’s content but it may not be as effective or have the same return on investment as other strategies.
Organisations should decide whether to use provenance as an approach to countering the challenges they face. Those choosing to use provenance technologies will also have to consider how to implement them.
4.1.2 Introduction of provenance in content lifecycle
Organisations can have a lot of content. Some of this content is publicly available. Other content, such as drafts, may not be publicly available now but will become public in the future.
The content may be at various stages of update and editing in preparation for publication. It may be distributed across a variety of systems and may be subject to changes by many individuals.
Organisations may also have some content that they never intend to make public. Some content may pose challenges or risks to the organisation itself. As a result, organisations may choose strong provenance measures only for some types of content. They may also choose to protect content at the point of publication rather than at point of creation.
4.1.3 Timeframe for content verification
The public’s requirements for information verification can vary in timeframe depending on the information context. Some information verification requirements will be aimed at short-term concerns such as elections, while others will be aimed at generational issues such as evidence concerning distant historical events.
For short-term events, the organisational risk is that it will take longer to verify the provenance information than the event timeframe requires. Timeframe issues can impact how long provenance records must be maintained, as well as how readily-accessible the records need to be.
4.1.4 Cost
Digital provenance mechanisms are relatively new and have associated implementation, operation, and maintenance costs. In most cases, organisations will have to change business processes to make effective use of provenance mechanisms. In addition, provenance technologies are evolving rapidly, and near-term implementations may quickly become obsolete.
Organisations may choose to prioritise non-provenance public information trust responses or they may choose to implement interim or partial solutions, for example using public provenance measures only for critical content.
4.1.5 Audience and format
The audience for provenance information may not necessarily be the same as an organisation's core audience. This will depend on an organisation's strategic and tactical response to the use of their information.
Formats for provenance information will be different depending on the system used by the specific audience.
Media companies have copyright on their information and may be able to use copyright tools to remove infringing material from the internet. In this case, the audiences for provenance evidence are legal professionals, Internet Service Providers and social media companies. Provenance information would need to be formatted to meet their different evidence requirements. A media company's implementation of provenance mechanisms will likely differ from that used by organisations whose provenance information audience is the public.
4.1.6 Maturity of public provenance technologies
Organisations should also consider the maturity of public provenance technologies. Technologies for versioning and logging to meet an organisation’s internal provenance requirements are mature. Public provenance technologies are less developed, although some of the related technologies used in private provenance, such as cryptographic hashing, can be used in public systems.
Publicly accessible provenance systems have additional requirements, for example, end-point devices such as cameras that can cryptographically sign content, and tamper-proof ledgers. These technologies are developing, but immature.
Organisations may choose to do partial and trial implementations. They may also choose to establish architectures that allow newer technologies to be integrated as they become available.
4.2 Example use cases
As we have seen, requirements for public provenance will vary between organisations depending on the challenges they face in communicating facts to their audiences and their public information trust strategy and tactics. These different requirements will shape the provenance infrastructure.
Here we provide analysis of 5 different use cases, using the provenance characteristics identified in Section 3.1 What to consider when selecting provenance systems.
An organisation that wishes to establish provenance of its own public content can either:
- establish a provenance record, including date and time, at time of publishing, or
- create content provenance records for all the intermediate steps of creating the content
The provenance record then becomes a tool for the organisation’s communications staff, as well as for others who review and provide fact-checking on the content, to validate or refute content veracity claims.
The public needs to be able to find and verify content easily.
To be useful, the verification mechanisms must be simple, intuitive, and reliable.
The required duration of a provenance record can vary depending on its expected use. Like many forms of digital record, some provenance records may only be required for a relatively short period, for example content that is transitory and only has short-term significance, such as event announcements. Provenance on the announcement may have value prior to the event, but the value of any provenance record will rapidly diminish afterwards.
Provenance infrastructure that supports short-duration requirements would not need to factor in long-duration requirements, which simplifies implementation and lifecycle considerations.
Some organisations will need their provenance records to endure well into the future. First-hand accounts of noteworthy events are one example. Future generations may need to verify the authenticity of today’s digital content. This is especially true in a world where generative AI is increasingly capable.
Proving the veracity of recorded testimony in timeframes of over 25 years could be challenging as certification components for identities and timestamps may not endure. The provenance mechanism must therefore address changes in technology, as well as turnover of business entities such as certificate providers and hosting services. The verification mechanism itself must also endure.
Maintaining the provenance and verification mechanisms over the long term may have to rely on distributed content stores and ledgers given that most organisations, along with their infrastructure, will eventually shut down as part of normal organisational lifecycle. Such mechanisms are still in the early stages of development and can be expensive to implement and use.
Some provenance requirements have privacy and anonymity considerations, for example in the field of journalism, where sources working in dangerous environments may need to remain anonymous for their protection. This can be done using trusted anonymous identities for individuals, or trusted capture devices that preserve user anonymity. Although this diminishes the strength of the provenance claim it can still add value.
Other provenance methods such as trusted timestamps and provenance certification by higher level entities (in this case, the journalist organisation) can strengthen the provenance record, helping to retain its usefulness.
Public content provenance records can potentially be used by organisations in their efforts to redress copyright infringement of their content.
The provenance mechanism can be used to identify copyright permissions available to others using the content (for example, Creative Commons licence) in a way which the public can verify.
Many jurisdictions are currently developing mechanisms to address other forms of information misuse.
Organisations implementing provenance mechanisms for this purpose may need to consider both specialised audiences and legal redress requirements in their systems design.