Digital trust and safety has many challenges in the digital services ecosystem. In the absence of standards, organizations and agencies have adopted their own, often clashing, versions. But it’s vital that an international standard be set so that companies and governments can protect universal human rights, speak the same “language,” and drive tangible progress in this space.

To address this, the Digital Trust & Safety Partnership (DTSP) created The Safe Framework. Joining Shane on Explain to Shane to discuss this and more are David Sullivan and Farzaneh Badiei. David serves as the executive director of DTSP and was previously co-chair of the Digital Safety Risk Assessment Framework at the World Economic Forum’s Global Coalition for Digital Safety. Farzaneh is head of outreach and engagement at DTSP. Both bring their extensive experience in trust and safety for an interesting discussion.

Below is a lightly edited and abridged transcript of our discussion. You can listen to this and other episodes of Explain to Shane on AEI.org and subscribe via your preferred listening platform. If you enjoyed this episode, leave us a review, and tell your friends and colleagues to tune in.

Shane Tews: What is the difference in trust and safety versus cybersecurity or content moderation, or are they complementary to each other?

Farzaneh Badiei:Trust and safety refers to the part of a digital services operation that focuses on understanding and addressing harmful content or conduct associated with that service. It’s much broader than content moderation, and content moderation can be a part of trust and safety. But when we talk about trust and safety as an industry term that must be understood as a bundle. 

You cannot have trust and not have safety. And if you focus only on safety and ignore trust, then you’re creating platforms where content is taken down arbitrarily, where civil society doesn’t know why access to services is being restricted, and people could lose confidence in expressing themselves on platforms and accessing services.

But it can also go the other way, too. So, a platform that’s technically safe, but users don’t trust it because it’s not serving anybody, and there are no accountability mechanisms. The difference is that really trust and safety is a kind of framework of holistic best practices, from product development to the enforcement of the rules of the platform.

David Sullivan: Especially today, there are a lot of misconceptions about what trust and safety mean. And people believe that we don’t need trust, we just need safety. So, this term, trust and safety, it goes back more than 20 years. The first companies that had trust and safety teams in the relatively early days were companies like eBay.

And the trust was about, can you trust the information so that you know, let’s say that this seller that you’re buying from is reliable. The trust is really about the integrity of the information that the sites are providing and ensuring that people trust the site to be able to use it. So that’s why trust and safety is this compound term that’s so important.

Tell us about The Safe Framework that DTSP put together.

David Sullivan:The framework is really about trying to create an overall kind of understanding of what it is that companies should be doing when it comes to user-generated content and user behavior on websites, and how to prevent the kinds of risks that everybody’s concerned about. And so, really, we found there are five overarching commitments that we organize all our work around, which is the foundation of that Safe Framework. 

The first is around product development, the idea of safety by design, of making sure that when you’re building a product, you’re thinking about keeping it safe from the outset and that you’re sort of going through the different practices that are going to help ensure that you get to a good outcome where you haven’t neglected to put in place the kinds of things you would need to make sure that a given product is as safe as possible, recognizing that we live in a world where you can’t ever be 100 percent safe.

In this framework, you have five fundamental commitments that you have ultimately landed on. Walk us through those.

David Sullivan: Earlier, I mentioned product development and safety from the outset. The second is around governance and making sure that, whatever the terms of service, community guidelines, content policies, whatever term it is for what is allowed and what’s not allowed on the site, is well thought through, clearly explained, takes in the perspectives of users, and is transparent and accountable. 

The third is about the enforcement of the governance and making sure that that enforcement is effective as well as proportionate. Improvement over time is the fourth. And then the fifth is transparency. And transparency isn’t just about issuing a report. It can be about providing notice to users. It can be about working with academics and researchers. So those five commitments are intended to kind of cover everything. 

And then underneath those five commitments, we have about 35 examples of best practices that we’ve set out. Those best practices are not exhaustive or exclusive. There may be other practices. This is a framework that we want to evolve over time, especially as technology develops. But those are examples of the types of things companies can do to fulfill the commitments.

Farzaneh Badiei:We are not a multi-stakeholder organization; we are an industry consortium, but we do a lot of outreach with civil society. And as David said, practices are not stagnant; they evolve. And for them to evolve, we need to get the feedback from civil society and other stakeholders. We also have the best practices, which are content and conduct-agnostic. They are not about issue-specific matters, like disinformation and stuff like that. But we do have, we do provide guidelines from time to time about issues such as age assurance. We have an age assurance guideline report, and also, there are guidelines about AI and automation of content and moderation.

And another piece that we are working on at the moment is the out-of-court dispute settlement bodies that were formed under EU law, and companies are trying to come up with some kind of guidelines about that as well. The consultation about that is open at the moment since it’s kind of like an iterative work that we are working towards.

Can you tell is a little bit about a goal that you had in creating this framework?

Farzaneh Badiei:I’d like to highlight the importance of speaking a kind of baseline language. It doesn’t have to be necessarily the same, but we should be able to understand each other when we say that, “Oh, we need to use blocking in this circumstance, or we need to use filtering.”

So what does that mean in the world of engineers who are doing this? And what does it mean in the world of policymakers, whether we have a baseline understanding? And it is very important for civil society as well, because if civil society wants to hold platforms accountable and to their word, they should have a good understanding of what these terms mean when they’re using their day-to-day operation. 

And for standards, I want to believe that standards can actually bring that level setting and being on the same page. And I’m very excited about this, the glossary being standardized, and hopefully even our scholars and academics can speak on safe terms, similar terms.