select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
Scientists comment on news that Meta has settled a court case and agreed to pay billions of dollars and implement changes to their platforms for teenagers.
Prof Andy Miah, Chair in Science Comms & Future Media, University of Salford, said:
“The case against Meta conflated two critical concerns about child online safety and treated them as though they were part of one problem. Exposure to harmful content and the habitual consumption of content are both risks, but they needed separating much more clearly in this case to ensure the recommendations will be effective.
“The new measures will send a shockwave through the digital and tech sector, the outcome of which will be a lot more attention to social responsibility, but there is a huge amount of misunderstanding in all of this. Behaviours that we loosely call addictive are, often, far from what we call addictive in a medically defined sense and, if we treat them as if they are the same, we lose insight into the problem.
“Putting in bans and curfews may sound like a strong principle and will be very popular, but it’s a blunt instrument that doesn’t take into account how young people interact today with every aspect of their lives. While other social media companies are likely to follow, it needs a much more rigorous societal response with digital wellbeing discussions in schools and a more positive conversation about the power of digital to do good. The impact of telling children they can’t have something, without a positive conversation about what else could be done will frustrate the outcomes.
“Making it easier for parents to understand what their children are seeing is where I’d place the emphasis, but this is useful especially for younger children. As we get into the 13-16 year old age range, we need a much more sophisticated approach that isn’t easily solved by just bans and curfews”.
“The advent of artificial intelligence will make this ruling seem like a drop in the ocean of behaviour change that is happening as a result of digital transformatons. We need to adopt the mindset that social media isn’t what it used to be and turn our attention to how socialisation through artificial intelligence is ushering in a new era of potential harms for young people.”
Dr Rachael Kent, Senior Lecturer Digital Economy & Society Education, King’s College London, said:
“This case reflects an important regulatory shift: from asking individuals to manage their digital wellbeing towards holding platforms accountable for the design mechanisms that drive and optimise addictive and harmful engagement.
“Importantly, some of the measures being introduced do not remove many of the underlying engagement mechanisms altogether – infinite scroll, autoplay and algorithmic recommendation remain – but introduce limits and stopping points around them. So this is a meaningful intervention, but it does not fundamentally redesign the engagement-driven model.
“There is growing evidence associating intensive and problematic social-media use with sleep disruption, anxiety, body-image concerns, compulsive or addictive use and displacement of exercise, education and face-to-face interaction [1,2]. The harm is not simply about time spent online, but what young people are repeatedly exposed to and how algorithmic systems [3] can amplify harmful or emotionally provocative content to sustain engagement.
“My understanding is that Meta’s new safeguards are US-wide jurisdictional requirements arising from this settlement, rather than a confirmed worldwide rollout. Technically, daily limits and nighttime blocks are relatively straightforward to implement; effective age assurance and circumvention are harder.
“These interventions could certainly help, particularly by protecting sleep and introducing stopping points into otherwise continuous experiences. Importantly, they also change the choice architecture rather than expecting teenagers to continually self-regulate against systems optimised for engagement and addictive use.
“But they only go so far. Time limits do not fundamentally address how recommendation algorithms select and amplify content, or the wider commercial incentives to maximise attention. Digital wellbeing cannot rest solely on individual responsibility.”
References:
1. Kent, R. (2020). Self-Tracking Health Over Time: From the Use of Instagram to Perform Optimal Health to the Protective Shield of the Digital Detox. Social Media + Society, 6(3)
2. The Digital Health Self, Wellness, Tracking and Social Media by Rachel Kent
3. Kent, R. (2024). (In) visibility of health and illness: Instagram as an unregulated public health platform. Anthropology & Medicine, 31(3), 297–312. https://doi.org/10.1080/13648470.2024.2386887
Prof Alan Woodward, Professor of Cybersecurity, University of Surrey, said:
On what the settlement does and does not establish
“This is a settlement, not a verdict. Meta has agreed to pay up to $16.7 billion and to change how teenagers use Instagram and Facebook. It denies the allegations, and no court has found that its products harmed anyone.
“The states alleged that Meta built features to keep young people scrolling and then kept quiet about what its own research showed. Those claims will now never be tested before a judge. What we have is Meta’s internal research as characterised by the states, plus the testimony of former staff. That is not the same as a legal finding.”
On the features at issue
“The features are not exotic. Infinite scroll and autoplay remove the natural moment where you might put the phone down. Notifications arrive unpredictably, which is the pattern of reward that keeps people checking. Like counts and filters feed social comparison.
“None of that is secret engineering, and all of it is standard across the industry. The scientific question is whether it does measurable harm to teenagers.”
On whether British teenagers will see the changes
“The changes are legally binding only in the United States. Meta usually ships teen safety features worldwide once it has built them, and European regulators are pressing for similar measures, so some of this may reach British teenagers.
“Nobody should assume it will. Hard limits cost the company engagement, and it has no legal obligation to accept that cost outside the US.”
On how hard the changes are to implement
“Building a timer or an overnight block is easy. Meta already has both in softer forms. The hard part is knowing who is a teenager.
“Every limit in this agreement applies only to accounts the company identifies as belonging to a minor, so the whole scheme rests on age checks. Those checks are imperfect. A child with a false date of birth or a parent’s login is invisible to them.
“China imposed playtime limits on minors’ gaming in 2019 and tightened them in 2021. When researchers examined around seven billion hours of play data, they found no credible evidence that heavy gaming by young people fell, partly because minors used adult accounts [1].”
On whether the interventions will work
“Whether daily caps and overnight blocks improve teenage mental health is untested at this scale. The most cited trial limited 143 undergraduates to 30 minutes of social media a day and found reduced loneliness and depressive symptoms after three weeks [2]. That is encouraging, but it is one small study over a short period, and it has not been replicated at scale.
“What is different here is that the limits are defaults rather than optional settings. Defaults are one of the most reliable tools in behavioural science. Countries where organ donation is the default have consent rates above 90 per cent, while otherwise similar countries that require opting in sit below 30 per cent [3].
“The settlement also requires an independent auditor to assess whether the measures work. If those results are published, they will tell us more about the effect of design limits on adolescents than anything in the current literature.
“Until then, the summary is that these mechanisms are plausible, the sleep evidence is strongest, but proof that time caps make teenagers healthier does not yet exist.”
References:
Dr Holly Bear Senior Postdoctoral Researcher, Department of Psychiatry, University of Oxford, said:
“It’s good to see action being taken on this urgent issue. Our recent research [1] found that one in three adolescents had encountered harmful content online in the past month, and of those exposed, almost three quarters encountered it entirely unintentionally, most commonly through algorithmically curated feeds, rather than by searching for it themselves. That tells us platform recommendation systems are likely driving a substantial proportion of harmful content.
“Of the specific measures announced, non-personalised feeds are a plausible idea, since engagement-optimised algorithms tend to push more extreme or emotionally charged content, but independent evidence testing this specific measure is thin, since researchers are rarely given access to evaluate platform changes directly.
“Measures like this are an important first step, but they won’t be enough on their own. Adolescents move across many connected platforms and features, so real progress needs action across the wider digital environment, not just changes at one company.”
Reference:
1. Bear, H., Soneson, E., Geulayov, G. et al. Algorithmically driven exposure to online self-harm content and its association with broader psychosocial risk in 32,000 adolescents. Nat. Mental Health (2026).https://doi.org/10.1038/s44220-026-00682-w
Declared interests
Prof Andy Miah:None
Dr Rachael Kent: She recently won a first-of-its-kind lawsuit against the technology giant Apple. Representing 36 million UK consumers, the case found that Apple had abused its dominant position through its App Store practices, including charging commission rates of up to 30%. The ruling paves the way for £1.5 billion in compensation to be returned to UK consumers and businesses, and sets an important precedent for challenging monopolistic practices in digital markets.
Dr Kent’s first book, The Digital Health Self: Wellness, Self-Tracking, and Social Media (Bristol University Press, 2023), offers a major analysis drawing on more than 15 years of research into how digital technologies structure wellbeing practices, self-tracking, and the pressures of self-optimisation.
Alongside her academic work, she is the founder of Dr Digital Health, through which she advises businesses, the NHS, government, and policymakers on the harms of digital technology and strategies to support healthier digital habits. She also hosts Digital Health Diagnosed: Your Dose of Tech Wellbeing, an evidence-based podcast on technology and health, featured on ITV News.
Prof Alan Woodward: None
Holly Bear: has provided paid consultancy to Girl Effect, a non-governmental organisation focused on improving the health, education and livelihoods of girls, arranged through Oxford University Innovation.