Risk Assessment for TalkMH

Welcome to TalkMH

TalkMH is a safe, supportive community where you can be heard, be yourself, and find the support you need.

Safe & Inclusive Everyone is welcome

Private & Confidential Your privacy matters

Judgement Free Be yourself, always

Supportive Community We're in this together

TalkMH Online Safety Risk Assessment

This risk assessment identifies foreseeable online safety risks on TalkMH and the controls used to reduce harm.

Service: TalkMH
Type: User-to-user mental health discussion forum
Intended audience: Adults aged 18+; operated from the United Kingdom
Assessment version: 2.0
Review frequency: Every 6 months, or sooner where a review trigger applies.

Purpose

TalkMH exists to provide peer support and discussion around mental health. It is not a medical service, crisis service, therapy provider, safeguarding authority, or emergency response service.

User Groups

Adults seeking peer support for mental health difficulties.
Adults discussing anxiety, depression, trauma, loneliness, relationships, grief, addiction, or stress.
Potentially vulnerable adults experiencing crisis, isolation, emotional distress, or suicidal thoughts.
Bad actors attempting spam, exploitation, scams, harassment, grooming, or harmful encouragement.

Overall Risk Rating

Because TalkMH is a mental health forum, the overall risk level is assessed as Medium to High. The highest-risk areas are suicide/self-harm content, vulnerable user exploitation, harassment, illegal content, and crisis disclosure.

Priority Illegal Content Assessment

Current Ofcom guidance requires priority illegal-content harms to be considered separately. This table records the likelihood, potential impact, overall risk, key controls and residual risk for each priority category relevant to the current guidance.

Priority harm	Likelihood	Impact	Risk	Controls / rationale	Residual risk
Terrorism	Low	Very High	Medium	Illegal/extremist content prohibited; reporting and prompt moderation; serious cases escalated where legally required.	Low
Child sexual exploitation and abuse (CSEA)	Low	Very High	High	18+ intended audience; sexual exploitation prohibited; critical handling procedure; statutory NCA reporting where applicable.	Low to Medium
Grooming	Low	Very High	Medium	Predatory conduct prohibited; reported private conduct reviewable; suspected under-18 accounts restricted; safeguarding escalation.	Low
Image-based child sexual abuse material (CSAM)	Low	Very High	High	Strict prohibition; immediate restriction; avoid unnecessary redistribution; statutory reporting process where applicable.	Low to Medium
CSAM URLs	Low	Very High	High	Illegal links prohibited; prompt removal/restriction; statutory reporting process where applicable.	Low to Medium
Hate	Low to Medium	High	Medium	Identity-based abuse prohibited; moderator removal; suspension or ban for serious/repeated conduct.	Low
Harassment, stalking, threats and abuse	Medium	High	High	Harassment/threats prohibited; report and moderation tools; content removal, thread locks and account restrictions.	Medium
Controlling or coercive behaviour	Low to Medium	High	Medium	Manipulation, threats and exploitation prohibited; reported public/private conduct reviewed; predatory accounts restricted.	Low to Medium
Intimate image abuse	Low	High	Medium	Non-consensual intimate imagery prohibited; prompt removal; serious cases escalated where appropriate.	Low
Extreme pornography	Low	High	Low	Explicit illegal sexual content prohibited; removal and account action when identified.	Low
Sexual exploitation of adults	Low to Medium	High	Medium	Predatory/exploitative conduct prohibited; privacy warnings; report/review and account restriction.	Low to Medium
Human trafficking	Low	Very High	Low	Illegal exploitation/recruitment prohibited; suspicious activity removed and escalated where legally required.	Low
Unlawful immigration	Low	High	Low	Illegal facilitation prohibited; reported unlawful content removed or restricted.	Low
Fraud and financial services offences	Medium	Medium to High	Medium	Scams/phishing/impersonation prohibited; suspicious links removed; spam and abusive accounts restricted.	Low to Medium
Proceeds of crime	Low	High	Low	Illegal transactions and facilitation prohibited; suspicious criminal activity removed and escalated where appropriate.	Low
Drugs and psychoactive substances	Low to Medium	High	Medium	Peer recovery discussion distinguished from illegal sale/supply; sourcing or criminal facilitation prohibited.	Low
Firearms, knives and other weapons	Low	High	Low	Illegal sale/supply/facilitation prohibited; reported illegal offers or instructions removed.	Low
Encouraging or assisting suicide or serious self-harm	Medium	Very High	High	Dedicated policy prohibits methods, instructions, encouragement, pressure, pacts and graphic content; prompt moderation and crisis signposting.	Medium
Foreign interference	Low	High	Low	Service purpose/scale makes this unlikely; illegal coordinated activity prohibited and removable.	Low
Animal cruelty	Low	Medium to High	Low	Illegal harmful content prohibited; reported material removed and accounts restricted where appropriate.	Low
Cyberflashing	Low to Medium	High	Medium	Explicit sexual content/sexual harassment prohibited; reported private communications can be reviewed; content/account action available.	Low to Medium

TalkMH-Specific Operational Risks

The following table records additional service-specific risks arising from TalkMH's mental-health peer-support model. These operational risks supplement, rather than replace, the priority illegal-content assessment above.

Risk	Likelihood	Impact	Rating	Controls	Residual Risk
Suicide or self-harm encouragement Users may post methods, encouragement, suicide pacts, countdowns, graphic descriptions, or content that assists harm.	Medium	High	High	Rules prohibit methods, instructions, encouragement, pressure, suicide pacts and graphic content. Report button available on user content. Moderators remove harmful self-harm content promptly. Crisis guidance displayed publicly. Repeat offenders are restricted or banned.	Medium
Immediate crisis disclosures A user may state they are about to harm themselves or someone else.	Medium	High	High	Clear disclaimer that TalkMH is not an emergency service. Users are directed to 999, NHS 111 and Samaritans. Moderators avoid counselling or diagnosing. High-risk posts are escalated internally for urgent review. Where lawful and necessary, serious threats may be reported to authorities.	Medium
Illegal content Users may post, request, link to, or promote illegal content.	Low to Medium	High	High	Rules prohibit illegal content including threats, terrorism, fraud, CSAM, hate crime and non-consensual intimate images. Illegal content is removed or hidden promptly. Accounts posting serious illegal material are banned. Serious unlawful content may be escalated to law enforcement or relevant agencies. Moderator actions are logged.	Medium
Child sexual abuse material or child exploitation Although TalkMH is adults-only, users may attempt to upload or link to illegal child sexual abuse material.	Low	Very High	High	18+ rule clearly stated. Sexual content involving anyone under 18 is strictly prohibited. Suspected CSAM is not shared internally except as necessary for reporting. Content is preserved only where legally appropriate and handled to avoid unnecessary further dissemination. Detected and previously unreported CSEA content is reported to the National Crime Agency where the statutory reporting duty applies. Offending accounts are immediately banned.	Low to Medium
Harassment, bullying or dogpiling Users may target vulnerable members through abuse, intimidation, stalking, ridicule or repeated hostile replies.	Medium	Medium to High	Medium	Rules prohibit harassment, threats, stalking, dogpiling and targeted abuse. Report function available. Staff may lock threads, remove replies, warn users or suspend accounts. Repeat offenders are escalated to longer restrictions or bans.	Low to Medium
Hate speech or discriminatory abuse Users may attack others based on protected characteristics or identity.	Low to Medium	High	Medium	Rules prohibit hate speech and identity-based abuse. Moderators remove hateful content. Severe or repeated breaches result in suspension or permanent ban. Potential hate crime may be escalated where appropriate.	Low to Medium
Exploitation of vulnerable users Bad actors may manipulate distressed users for money, attention, images, personal details or emotional dependency.	Medium	High	High	Rules prohibit scams, manipulation, impersonation and asking users for money or personal information. Users are warned not to share private details publicly. Suspicious behaviour can be reported to staff. Private message concerns are reviewed where users report them. Predatory accounts are banned.	Medium
Non-consensual intimate imagery Users may post intimate images, revenge porn, AI-generated sexual images, or private screenshots.	Low	High	Medium	Rules prohibit explicit sexual content and non-consensual intimate images. Content is removed promptly when identified. Accounts may be banned immediately. Serious cases may be escalated to authorities.	Low to Medium
Privacy breaches and doxxing Users may share real names, addresses, workplaces, screenshots, social profiles or private messages.	Medium	Medium to High	Medium	Rules prohibit sharing personal or identifying information without consent. Moderators remove doxxing and private screenshots. Users are encouraged to avoid posting personal details. Repeat or malicious privacy breaches lead to restrictions or bans.	Low to Medium
Medical misinformation or harmful advice Users may give unsafe advice about medication, diagnosis, treatment, stopping medication, or replacing professional care.	Medium	Medium to High	Medium	TalkMH states it is peer support only. Users are discouraged from diagnosing others or giving medical instructions. Posts encouraging dangerous medical action may be removed. Users are signposted to qualified professionals and crisis support where appropriate.	Medium
Spam, scams, phishing and malware Users may post unsafe links, fake support services, phishing pages or malicious downloads.	Medium	Medium	Medium	Rules prohibit spam, unsafe links, phishing and malware. New users may be manually reviewed or restricted. Suspicious links are removed. Spam accounts are banned. Forum anti-spam tools are enabled where available.	Low to Medium
Under-18 access Minors may attempt to access an adults-only mental health forum.	Medium	Medium to High	Medium	TalkMH is clearly marked as 18+. Registration requires confirmation of adult status; self-declaration is not treated as highly effective age assurance. Accounts suspected to be under 18 may be restricted or removed. A separate Children\'s Access Assessment is maintained and reviewed in accordance with current Ofcom requirements. Rules prohibit sexual content and exploitative behaviour.	Medium

Children's Access Assessment

TalkMH is intended for adults aged 18+, but an age statement or registration self-declaration is not treated as highly effective age assurance. A separate Children's Access Assessment must therefore be maintained and reviewed in accordance with the Online Safety Act and current Ofcom guidance.

If TalkMH is assessed as likely to be accessed by children, the applicable Children's Risk Assessment and protection-of-children duties must also be completed.
If highly effective age assurance is introduced, the assessment must be updated to reflect how it is implemented and evidenced.
The Children's Access Assessment should be reviewed at least annually, and sooner where required.

Required Controls

Visible community rules available before and after registration.
Report button available on user-generated content.
Moderation queue or manual review for new users where possible.
Clear suicide/self-harm policy.
Clear illegal content policy.
Moderator procedure for urgent risk, illegal content and repeat offenders.
Incident log for serious reports and moderation actions.
Process for reporting detected and previously unreported CSEA content to the National Crime Agency where the statutory duty applies.
Separate Children\'s Access Assessment and any consequential Children\'s Risk Assessment where required.
Record of applicable Code of Practice measures and the rationale for any alternative measures.
Appeals process for moderation decisions.
Regular review of rules, reports, incidents and emerging risks.

Moderator Response Levels

Level	Example	Response
Low	Off-topic post, mild argument, accidental rule breach	Reminder, edit, move, or informal warning
Medium	Harassment, repeated hostility, unsafe advice, privacy issue	Remove content, formal warning, thread lock, temporary restriction
High	Threats, self-harm encouragement, doxxing, predatory behaviour	Immediate removal, account restriction, senior moderator review
Critical	CSAM, terrorism content, credible immediate threat, serious illegal content	Immediate action, preserve necessary records, escalate where legally required

Record Keeping and Evidence

A written copy of each completed or revised risk assessment is retained.
Serious moderation incidents and legally significant escalations are recorded with the date, action, decision-maker and outcome.
Evidence supporting risk ratings is retained where reasonably available, including moderation/report trends, service changes and relevant Ofcom risk-profile information.
Where an alternative measure is used instead of an applicable Code of Practice measure, the alternative and the basis on which it is considered compliant are recorded.
Records are maintained in a form capable of being provided to Ofcom if lawfully requested.

Review Triggers

A serious self-harm, suicide, abuse, exploitation, or illegal content incident.
A significant increase in reports or moderation workload.
Launch of new features such as private messaging, images, groups, live chat, or anonymous posting.
Changes to UK law, Ofcom guidance, Codes of Practice, Risk Profiles, categorisation or enforcement expectations.
Evidence that existing controls are not working effectively.
Before any significant change to the design or operation of the service that may affect online-safety risk.

Assessment owner / person accountable for online safety: TalkMH Community Team
Version: 2.0
Last reviewed: 7 August 2026
Next routine review due: 7 February 2027