[Your Website Name] — Online Safety Act Risk Assessment & Children’s Access Assessment

1. About our website and chat service

This document records our illegal content risk assessment and our children’s access assessment under the UK Online Safety Act 2023 for the World of Chat Messenger chat on our website. [If you concluded children are likely to access the service, add: It also records our children’s risk assessment (Section 3).]

• Website name, URL & topic: [Website name, URL and a brief description of what your site is about]
• Who uses the chat: [Open to logged-in members and guests, OR logged-in registered members only. Approximate number of monthly chat users.]
• What users can do: [e.g. Public chat, private messages, share photos (on/off), see who is online, block users, report abuse]
• What is kept (90-day auto-delete): Messages, photos and moderation logs are kept for 90 days so reports can be investigated, then automatically and permanently deleted. If we report child abuse content to the NCA, the reported content, report details and related user data (including two weeks of chat history) are saved separately and kept for 1 year, and the report reference number for 5 years, as the reporting regulations require.
• Separate sites: World of Chat runs the backend servers, but our site’s chat is isolated from every other website. Users on other sites cannot see or join our chat. We control who can use our chat and are the provider responsible for it.
• Our approach to safety: [1–2 sentences on your rules and moderation, e.g. Our rules prohibit illegal content, pornography, harassment and scams. Reports are reviewed daily and rule-breakers are removed.]

2. Children’s access assessment (Online Safety Act, section 35)

We have assessed whether children (under 18s in the UK) are likely to access our service. We will repeat this assessment at least once a year, and sooner before any significant change to the chat or if we see evidence of more children using it.

• Q1. Is it possible for children to access the service? [Unless you use age verification or age estimation that means children are not normally able to get in, answer: Yes. We do not use age verification or age estimation, so we cannot conclude that children are unable to access the chat. An 18+ rule or self-declaration does not change this answer.]
• Q2a. Do a significant number of children actually use the service? [Give your evidence: your audience, what moderators and reports show, sign-up data, number of users removed for appearing to be under 18 in the last year.]
• Q2b. Is the service of a kind likely to attract a significant number of children? [Consider your topic, design and how people reach the chat, e.g. whether guests arriving from search engines can use it.]
• Our age policy (optional): [If you run an 18+ community, say so and how you handle users who appear underage, e.g. Our chat is for adults. Users who appear to be under 18 are removed and users are encouraged to report them. This is a safeguard; it is not age assurance.]
• Conclusion: [EITHER: The child user condition is not met, so our service is not likely to be accessed by children. OR: Our service is likely to be accessed by children, so we have completed the children’s risk assessment in Section 3.]

3. Protecting children from harmful content (complete if children are likely to access the service)

Content harmful to children includes primary priority content (pornography; content encouraging suicide, self-harm or eating disorders) and priority content (e.g. bullying, abusive or hateful content, serious violence, dangerous challenges).

• What our rules prohibit: [Confirm your terms ban pornography and suicide/self-harm/eating-disorder content, and list other harmful content you ban.]
• How harmful content is found and removed: [User reports with 15-message context, Chat Monitor, how often it is checked, how quickly content is removed.]
• Explicit photos: [Photo sharing on/off. If on: the server-side scanner blurs likely-explicit photos until a viewer chooses to uncover them, as an extra layer alongside our rules and moderation.]
• Bullying and abuse: [1-click blocking, reporting, timed and permanent bans.]
• Grooming and contact from strangers: [How you notice and act on adults targeting children; whether private messaging is limited for users you believe are children; removal and reporting of offenders.]
• Children’s risk levels: [Give a Low / Medium / High risk level for pornography, suicide/self-harm/eating-disorder content, and bullying/abuse, with a sentence of reasoning for each.]

4. Illegal content risk assessment (17 kinds of priority illegal content)

Below is our assessment of the 17 kinds of priority illegal content identified by Ofcom, plus other relevant illegal content. Each risk level (Negligible / Low / Medium / High) reflects how likely the harm is on our site and how severe its impact would be, given our features, users and controls. We will update it if real-world use shows otherwise.

Kind of illegal content	Risk level	Evidence, risk factors & how we mitigate it
1. Terrorism	[Negligible / Low / Medium / High]	[Extremist promotion or recruitment. Rules prohibition, removal, bans, police referral.]
2. Child sexual exploitation & abuse: grooming	[Negligible / Low / Medium / High]	[Ofcom flags private messaging, contact with strangers and guest/anonymous users as grooming risk factors. List which your chat has. Mention how grooming would be noticed, reporting with context, blocking, bans, and reporting to NCA/police.]
3. Child sexual exploitation & abuse: CSAM (images and URLs)	[Negligible / Low / Medium / High]	[Photo sharing on/off, scanner and blur, reporting, immediate removal, evidence preserved, report to NCA (and IWF/police).]
4. Hate	[Negligible / Low / Medium / High]	[Rules, reporting, blocking, bans.]
5. Harassment, stalking, threats & abuse	[Low / Medium / High]	[The most common issue in live chat. 1-click blocking, reports with 15-message context, timed/permanent bans.]
6. Controlling or coercive behaviour	[Negligible / Low / Medium / High]	[Pressure or abuse via private messages. Blocking, no location tracking in the messenger, account removal on request.]
7. Intimate image abuse	[Negligible / Low / Medium / High]	[Sharing intimate images without consent. Photo sharing on/off, scanner and blur, reporting, swift removal, bans.]
8. Extreme pornography	[Negligible / Low / Medium / High]	[Photo sharing on/off, porn banned in rules, scanner and blur, removal, bans.]
9. Sexual exploitation of adults	[Negligible / Low / Medium / High]	[Commercial sex solicitation or exploitation. Rules, spam filters, bans.]
10. Human trafficking	[Negligible / Low / Medium / High]	[Relevance to your site, rules prohibition, police escalation.]
11. Unlawful immigration	[Negligible / Low / Medium / High]	[Organising unlawful entry. Relevance to your topic, removal and bans.]
12. Fraud & financial offences	[Low / Medium / High]	[Scam, phishing and romance-fraud messages. Rate limits, duplicate-message filters, rules against solicitation, reports, bans.]
13. Proceeds of crime	[Negligible / Low / Medium / High]	[No wallets, payments or money transfer features in the chat.]
14. Drugs & psychoactive substances	[Negligible / Low / Medium / High]	[Buying/selling drugs in chat. Rules, reports, bans.]
15. Firearms, knives & other weapons	[Negligible / Low / Medium / High]	[No marketplace features, rules prohibition, removal, police referral.]
16. Encouraging or assisting suicide	[Negligible / Low / Medium / High]	[Rules, no algorithmic feed, moderator removal, signposting to support (e.g. Samaritans, 116 123).]
17. Foreign interference	[Negligible / Low / Medium / High]	[Coordinated state-backed disinformation. Small community, no viral feed, rate limiting.]
18. Animal cruelty	[Negligible / Low / Medium / High]	[Images or videos of animal cruelty. Photo sharing on/off, rules, removal, bans.]
Other illegal content (e.g. cyberflashing, threatening communications)	[Negligible / Low / Medium / High]	[Unsolicited sexual images, other offences likely in chat. Scanner and blur, blocking, reports, bans.]

Note: Ofcom counts grooming and CSAM together as one of its 17 kinds (CSEA), which is why the rows above go to 18. Assess both parts.

5. What we already do (safety & moderation controls)

• Terms & rules: [Where your rules/terms are published and confirmation that they prohibit illegal content, pornography, harassment, scams and spam, and explain how users are protected.]
• Moderation: [Who moderates (WordPress administrators are chat admins automatically; list any moderators you’ve appointed), how often reports are checked, and how quickly illegal content is removed.]
• Usernames: Usernames suggesting child abuse, terrorism or other illegal activity are removed and the account banned.
• Bans (Account, Guest ID & IP): Admins and moderators can ban users for 1 hour, 24 hours, 7 days, or permanently. Registered members are banned by account; guests by both Guest ID and IP address.
• Reporting & complaints: Every user can report a user or message with one click. Reports include 15 messages of context and go to our Chat Monitor. [Say how users can complain about a moderation decision, e.g. by email.]
• Blocking: Every user can block anyone with one click.
• Explicit photo scanner & blur: Photos sent in chat are checked by a server-side scanner; likely-explicit images are blurred until a viewer chooses to uncover them. Automated scanning is one layer alongside our rules and moderation.
• Rate limits & anti-spam: Built-in rate limits and duplicate-message filters limit bot flooding and repetitive link spam.

6. Weak points & honest limitations

• Reliance on reports and moderators: Harmful text chat is mainly caught when a user reports it or a moderator sees it. [How often you check the Chat Monitor, e.g. daily.]
• Grooming: Grooming can look like ordinary friendly chat before it turns harmful, so we rely on users reporting concerns and moderators acting quickly.
• No age verification: [e.g. We do not use age verification or age estimation, so we cannot be certain no under-18s are present. We address this through our rules, moderation, reporting and removal of users who appear underage.]
• Automated scanning limits: The photo scanner can miss some explicit images or flag innocent ones.
• Ban evasion: A banned guest could return using a VPN or different network. (If this becomes a persistent problem, we can switch off Guest Chat so only registered members can chat.)

7. Responsibility, serious cases & review record

• Who is responsible: [Name and role of the accountable person, and the contact email users and Ofcom can use]
• Serious illegal content: If child sexual abuse material or grooming is reported or detected, we will remove it, ban the offender, preserve the evidence, and report it to the National Crime Agency through the CSEA Industry Reporting Portal at the priority the regulations require. We registered on the portal on [Date]. Terrorism content will be removed and reported to the police. If a child is in immediate danger, we call 999.
• Review schedule: This assessment was completed on [Date] and will be reviewed by [Date + no more than 12 months], and before any significant change to how chat works on our website (e.g. enabling photo sharing or guest chat).