The Hidden Risks of Social Media:
What's Actually Happening on Every Platform
Before letting your child use a platform, create an account yourself and explore it as a user their age would. This guide tells you what to look for, and what you'll find.
Knowledge is the first step. But understanding is what changes things.
Ofcom Children's Online Experiences Report (May 2026) · NSPCC Online Harm Statistics (2026) · Imperial College London SCAMP Study
Every platform carries different risks. Every child needs different conversations.
The risks your child encounters on TikTok are not the same as those on Discord or Roblox. Generic advice about "social media" misses the specifics, and the specifics are where the real harm lives. This guide covers the platforms UK children are most likely to be using, with the primary harm pathways and the data behind them.
This isn't about banning platforms. It's about understanding them well enough to have conversations that actually matter. For the broader picture of how UK law is changing, read The Online Safety Act 2026.
Apps Every Parent Should Know
The stats below come from platform transparency reports, academic research, and regulatory filings. Where data is from earlier years, it reflects the most recent published figures available.
Before your child uses any platform: create a test account yourself using their age and spend 30 minutes exploring it. See the content the algorithm surfaces, the contact mechanisms that exist, and the design choices that keep users engaged. What you find will inform every conversation you have.
Snapchat
- Snap Map reveals real-time location to contacts, and potentially strangers
- Quick Add expands network beyond known contacts to strangers
- My Eyes Only hides content behind a passcode, parental checks won't see it
- Disappearing messages erase evidence of abuse, harassment, or grooming
- Explore and Reels push appearance-focused content to girls particularly
- DMs allow strangers to message children directly from public posts
- Suggested accounts expose children to fake profiles and predatory accounts
- Likes and followers create a public ranking system for social worth
TikTok
- Short-form For You Page is engineered to erode attention spans and maximise time on platform
- Algorithm can surface self-harm and eating disorder content within 2-3 minutes of new account creation
- DMs and live features enable grooming and direct contact with strangers
- Algorithm expands network into toxic communities without explicit consent
Roblox
- In-game chat allows strangers to contact children directly, including in private games
- User-generated games contain hidden adult content within child-safe-looking games
- Private servers enable predatory behaviour away from platform moderation
- Robux system creates financial pressure, children spend real money on in-game currency
Discord
- Unmoderated servers expose children to adult content with no platform oversight
- DMs expand network beyond the server into hidden one-to-one contact
- Screen share enables immersive, real-time predatory contact
- Frequently used to move children from TikTok and Roblox into private, unmonitored spaces
YouTube
- Shorts fragment attention spans, designed for rapid consumption and dopamine cycling
- Autoplay pushes children into increasingly extreme or age-inappropriate content
- Comments allow stranger interaction with children who post or comment publicly
- Livestreams enable direct, real-time predatory contact, particularly dangerous for younger children
Character AI
- AI companions blur the boundary between real friendship and simulated intimacy
- Role-play bots enable sexualised conversations that adapt to the user's inputs
- Custom bots can be specifically designed to target and manipulate children
- Bots can reinforce harmful behaviours, ideation, and isolation, invisibly and persistently
Telegram
- Large unmoderated groups expose children to extremist and radicalising content
- End-to-end encrypted chats enable covert grooming invisible to platforms and parents
- Network 764, a known child abuse network, uses Telegram as a primary coordination channel
- Disappearing messages reduce accountability for abuse and harassment
- Highly visual image feed amplifies body comparison and appearance anxiety, particularly in girls
- "Thinspiration" and pro-eating disorder boards persist despite platform policies, often framed as "wellness" or "fitness"
- Algorithm surfaces increasingly extreme body image content based on initial searches
- Pinterest paid a $7.5M FTC settlement for knowingly collecting data on children under 13
The Lasting Consequences of Unchecked Use
Beyond the immediate risks, unchecked social media use during critical developmental years has measurable long-term effects across four domains.
Cognitive Development
- Reduced attention spans from rapid content consumption
- Impaired critical thinking and problem-solving skills
- Weakened memory formation and recall abilities
- Disrupted executive function development
Mental Health
- Increased rates of anxiety and depression, especially in girls
- Body image issues and eating disorder risk
- Social comparison driving chronic low self-esteem
- Heightened risk of self-harm ideation through algorithmic exposure
Social Development
- Reduced face-to-face communication skills and confidence
- Difficulty reading non-verbal social cues
- Shallow relationships based on online validation rather than real connection
- Increased loneliness despite constant online connectivity
Physical Health
- Disrupted sleep, 60% of 11-17 year olds use phones midnight to 5am
- Sedentary behaviour from extended passive consumption
- Eye strain and posture problems from prolonged screen use
- Reduced time for the physical activity essential to development
The Developing Brain Is Particularly Vulnerable
The prefrontal cortex, responsible for impulse control and decision-making, doesn't fully develop until age 25. Meanwhile, platforms are designed by some of the world's most sophisticated engineers to capture and hold attention through variable reward patterns that mirror gambling psychology.
Research shows attention spans have dropped from 12 seconds to 8 seconds over the past two decades, and 70% of this decline is attributed to digital media consumption. This isn't a failure of character. It's the predictable result of a system designed to erode the very capacity children need to resist it.
For a deeper look at why this happens, and what parents can do, read Why Your Child Falls Apart After Screen Time.
Protection starts with awareness. Change starts with skills.
Signs to watch for
Five actions this week
Create accounts on the platforms your child uses
Spend 30 minutes exploring each as a user their age. See what the algorithm surfaces.
Say the most important thing explicitly
"If you see something that upsets you online, you will never get in trouble for telling me." Say it. Mean it.
Review privacy settings together
Check every platform, together, not for them. Explain what each setting does and why it matters.
Charge devices outside bedrooms at night
For every family member. Not a punishment, a structural protection. Shield addresses this.
Build a family tech agreement
Rules that are built together stick. Get the free Family Tech Agreement. Then take the Find Your Skills quiz to know where to start.
The six havyn skills that build lasting protection
Frequently Asked Questions
Which social media platforms are most dangerous for children?
All platforms carry risks, but the pathways differ. Snapchat and Instagram carry the highest sextortion risk. TikTok's algorithm can surface self-harm content within minutes. Roblox has significant contact risks for children under 13. Discord is frequently used to move children from other platforms into unmonitored private spaces. Character AI poses unique risks around emotional attachment and inappropriate role-play with AI bots.
What age should children be allowed on social media?
Most platforms set a minimum age of 13, though 81% of 10-12 year olds are already using them. The UK government is moving toward under-16 restrictions following the Children's Wellbeing and Schools Act 2026. More important than age alone is whether the child has the skills to navigate what they find, emotional regulation, critical thinking, and an understanding of how platforms work. Read What Is Digital Literacy? for the full picture.
How do I talk to my child about social media risks?
Start with curiosity, not surveillance. The most protective thing you can do is create a relationship where your child feels genuinely safe telling you when something goes wrong, without fear of punishment or devices being removed. Say this explicitly: "If you ever see something that upsets or confuses you online, tell me. You will never be in trouble for telling me." Say it more than once. Mean it.
What does the Online Safety Act mean for these platforms?
The Online Safety Act (in enforcement from 2026) requires platforms to implement age verification, make recommendation algorithms safer for children, and provide accessible reporting mechanisms. Ofcom wrote to TikTok and YouTube in May 2026 demanding safer feeds. But regulation sets a floor, it cannot build the skills children need. Read the full Online Safety Act guide.
What's the most effective thing I can do to protect my child?
Research consistently points to the same answer: the most protective factor is a relationship in which children feel safe to tell a parent when something goes wrong. Beyond that, building digital literacy skills, emotional regulation, critical thinking, healthy boundaries, gives children the internal resources to navigate social media safely. Take the free Find Your Skills quiz to find out where to start.
Awareness is the first step.
Skills are what changes things.
The havyn Challenge gives children aged 5-13 the six skills they need to navigate technology, one skill per week, starting July 2026.
Join the Challenge Find Your Skill First