The Gist: Here come Ireland's speech police

Ireland's regulators are getting ready to change what we can see and say on the Internet. This is the Gist.

The Gist: Here come Ireland's speech police
Photo by Jason Leung / Unsplash

Ireland's new regulator of Internet content took another step towards making large social media platforms scan our faces for biometrics or demand our government papers before allowing us to access them this week.

But that's not the only piece of Irish Government regulation of Internet content in the works. If we do a search in the EU Commission's TRIS website (regular readers will remember how TRIS works from Too Urgent to Get Right) we can see there are two far-reaching planned codes-Bills and proposed regulations- on internet speech notified from Ireland. Taken together, they may upturn many of the assumptions we have learned about using the Internet, not just in Ireland but also- as the Irish Regulations have effect on the huge tech platforms based here- across the EU.

Let's whizz through them in turn.

Part 5 of the Electoral Reform Act 2022

Oh, hey, did you know we already passed a Part 5 of the Electoral Reform Act two years ago that would have given the new Electoral Commission the power to decide they didn't like what people had said on a political matter and order it taken down? Have you also noticed you've never heard of that happening?

That's because it was never switched on (commenced). Probably because it was a sloppy piece of political showboating passed for rhetorical purposes without the rigour to actually make it work.

But now Part 5 is back, Back, BACK!

Here's what the Irish Times quoted from me on this bit of legislation:

The new draft, McGarr says, is better. “It’s more realistic, more defensible and less open to challenge in respect of breaching EU law – and it looks to create an actual operational system.”
He warns that much will rely on codes of conduct that have not yet been finalised: “They’ve punted all the really hard parts of this process down to codes of conduct that are not visible to us. It’s a framework, the actual rules are still unknown.”

Not quoted was my third thought on this:

This Act is a significant step up in operationalising what was previously little more than a legislative dead letter passed for rhetorical purposes.
Passing the new law will mean the state begins deliberately deciding what political speech- when such political speech is explicitly not illegal- is and is not allowed. This is a very major decision and therefore this is a significant Bill.

I'm not saying it is automatically bad. There is plenty of scope to decide certain deliberate misinformation is so undesirable that it could interfere with democracy. But what is notable is that even the framers of this legislation couldn't actually draft a set of rules to let us see what they had in mind.

With a vote on a new government expected before Christmas and this bill coming out of the TRIS standstill mechanism on the 4th October 2024, this one might get hot, fast.

If the Electoral Commission decides it would like to use its new powers and issues a Code of Conduct before the General Election, we're not going to have much time to debate whatever ideas it cooks up in the meantime.

💡
Link to TRIS Page and Proposed Text
Notification Number:2024/0374/IE (Ireland)
Date received:03/07/2024
End of Standstill:04/10/2024
Sponsoring Body: Department of Housing, Local Government and Heritage

Still open for public submission?: YES, here

Draft Online Safety Code

Background Gists to this particular doozie at Coimisiún na Meán go off the rails and Coimisiún na Meán is Big Mad.

After those Gists were published, the Commission received 1400 submissions on their plans, overwhelmingly negative. Having defended their hatstand proposal to age gate access to platforms entirely in press responses, they eventually stuck their actual response in a single sentence on page 17 in their report on the consultation:

"The Commission has removed the obligation in the Code to restrict the access of children to services through age verification."

But the draft Online Safety Code remains riddled with terrible ideas and evidence of just careless drafting. It still reads like a first pass at a document, but it's now come free of the EU standstill period (as of the 28th August 2024). That means this is The Text.

That text still includes the plan to require their regulated entities to collect ID data on children and adults seeking access to adult content and store it, creating a distributed "porn-user register" as the Coimisiún phrased it while denying it.

It also provides, at clause 10.9, an effort to address concerns about misusing these records (which will be stored by such stalwarts of probity and trustworthiness as Elon Musk's X)

a video-sharing platform service provider shall not process for commercial purposes (such as direct marketing, profiling and behaviourally targeted advertising) the personal data of minors collected or otherwise generated by providers [to comply with the requirements of the Code]

But in doing so, it opens up more questions than it answers. Are only the minors' records now sheltered from commercial use and advertising profiling? Will adults have a new little field distributed to behaviourally targeted advertisers letting them know what sort of adult content preferences they've accessed? Will both children's and adults' ID data be permitted to be processed for additional non-commercial purposes such as academic research (a means of laundering data processing in the field of AI model creation in the past). If neither party's data can be used in any of these ways, why create a special clause only covering children's data?

Will the day come where platform users look back in fondness to a time when the ads that followed them around the internet were just for slippers?

Despite this textual and conceptual shambles, the Coimisiún na Mean (Media Commission) can now go ahead and bring these restrictions in. They'll apply at first to their designated services (Facebook, Instagram, LinkedIn, Pinterest, Reddit, TikTok, Tumblr, Udemy, X, and YouTube) though they say other services may be added to that list.

What's interesting is that the commercial properties with the largest headcount of actual workers in Ireland on that list are also the least likely to be effected at first. There's a get-out for platforms who just make it a part of their terms and conditions that designated adult-only videos can't be uploaded or shared on their platform.

A video-sharing platform service provider whose terms and conditions do not preclude the uploading or sharing of adult-only video content as defined in this Code shall implement effective age assurance measures as defined in this Code to ensure that adult-only video content cannot normally be seen by children. An age assurance measure based solely on self-declaration of age by users of the service shall not be an effective measure for the purposes of this section.

Guess which large designated platform doesn't has rules about sharing nudity, even to the point of acknowledging that it does not apply these rules with nuance?

"the execution of our policies can sometimes be less nuanced than we would like and restricts content shared for legitimate purposes.",

Meta helpfully explains.

Who doesn't like avowedly 'less nuanced' regulation by the State and huge companies of the things they can and cannot say to their friends online?

Coda: This one looks like another candidate for the pre-election rush to regulate the internet.

💡
Link to TRIS page and Proposed Text
Notification Number:2024/0283/IE (Ireland)
Date received:27/05/2024
End of Standstill:28/08/2024
Sponsoring body: Coimisiún na Meán

Still open for public submission?: No.

Clouds on the horizon

Legislation and state systems to control what people say and what may be shared are frequently commenced with good intentions and then instantly degrade into abuse by power. These proposals will require expensive and lengthy court challenges, over and over again, if they start to illegitimately restrict speech.

It is easy to point to the things which are obviously harmful online. There is such a class of content (material promoting eating disorders to children, for example). The difficulty comes where the lines are to be drawn on the edges of those definitions. But when it comes to statute, what we've seen is the government leaving the drawing of those difficult lines to regulatory bodies and, so far, the regulatory bodies not being up to the job.