Ofcom investigates whether Meta risk-assessed Instagram's disappearing Instants before launch

A teenager takes a photo on Instagram and sends it to her close friends list. Once it has been viewed, it is gone ^[3]. Meta launched Instants in May as "a new way to share photos in the moment" ^[3]. According to Ofcom, Meta had to finish one piece of work before the first of those photos was sent.

The probe will examine whether Meta failed to carry out "a suitable and sufficient illegal content risk assessment and children's risk assessment before the launch of Instants" ^[2]. According to the BBC, the regulator said Meta had not properly assessed the feature's risks before launch ^[17]. George Lusty, Ofcom's enforcement director, said that "significant changes to platforms must be risk assessed before they're launched" ^[7]. "You wouldn't launch a new car without testing it properly, and it's the same for online services," he said ^[8]. "Safety by design must be built in, not bolted on," Lusty added ^[9].

Product teams tend to tell themselves that a regulator who has been briefed has more or less signed off. Meta's spokesperson said the company "conducted a risk analysis" and briefed Ofcom "on a number of occasions" before launch ^[5]. Ofcom opened an investigation anyway ^[1]. The duties the BBC describes are assessments that platforms must carry out and keep updated ^[10], and the test for this one is whether it was suitable and sufficient ^[2]. The BBC report does not say what Meta's analysis contained.

The type of feature matters too. The BBC reported that disappearing messages and images have previously been singled out as posing particular risks for children and teens on some apps ^[13]. The children's duties cover pornography and material promoting self-harm or suicide, classed as "priority primary content", alongside abusive, hateful or bullying content ^[11]. Meta points to built-in protections for Teen Accounts ^[12]. In my view those controls describe what the product blocks after a photo arrives. I'd expect a children's risk assessment to cover what can arrive in the first place, and from whom, in a format designed to disappear.

The probe comes while Meta, TikTok and X are in court arguing they should not have to provide so much information in response to Ofcom's formal requests ^[14]. Meta also argues that Ofcom should not be able to base fines on global annual revenue ^[15], and that its messaging and image-sharing apps should not carry extra duties such as tackling paid-for fraud ads ^[16]. I'd expect a dispute over whether an assessment was sufficient to be fought largely through those formal requests.

For a team with a UK launch on the calendar, I'd sort changes on two questions. One is whether the change alters who can see a piece of content. The other is whether it alters how long content lasts or whether it can be copied. A change that does neither goes into the next scheduled update of the existing assessments ^[10]. One yes gets a written check against both the illegal-content and the children's assessment before the release date. Two yeses get both assessments redone and finished before launch, with any briefing to Ofcom on top.

The cost is calendar time, because the release waits on the risk work. Teams will argue over Lusty's word "significant" ^[7] when they decide which box a change belongs in. Instants answers yes to both questions, with an audience of close friends and mutual follows ^[4] and images that vanish once viewed ^[3].