Harmful Online Content and Interpersonal Offences

Introduction and context

Digital communication has transformed how people form relationships, express opinions, organise communities and share information. The same technologies can also be used to intimidate, humiliate, exploit, threaten or silence others. Harmful online conduct ranges from repeated unwanted contact to the distribution of child sexual abuse material. Some conduct is clearly criminal; other conduct may be civilly actionable, subject to platform rules, or protected expression. The legal classification depends on the facts, the jurisdiction, the identity and age of the persons involved, the content, the intention or mental state of the accused, and the effect on the victim.

This chapter examines common interpersonal and content-based cyber-offences. It adopts a victim-centred and rights-conscious approach. Protection from abuse must be balanced with freedom of expression, privacy, due process and the prohibition of arbitrary censorship. Because criminal statutes differ, students should treat the legal categories discussed here as analytical frameworks rather than as a substitute for local legislation or legal advice.

Learning objectives

By the end of this chapter, a student should be able to:

1. Distinguish cyberstalking, harassment, threats, cyberbullying and related conduct.
2. Explain consent in relation to intimate images, communication and publication.
3. Describe sextortion, image-based abuse and child sexual abuse material.
4. Analyse hate-related offences and unlawful publication while respecting speech rights.
5. Apply a basic framework for victim protection, evidence preservation and reporting.
6. Identify the ethical and governance responsibilities of platforms, institutions and investigators.

1. A framework for analysing harmful online conduct

A useful analysis asks six questions:

The same message can have different legal consequences in different settings. A political criticism directed at a public official may be protected speech, whereas a credible threat of violence is generally treated differently. A consensually created image is not necessarily consent to its later distribution. A rude single comment may violate institutional policy without meeting the criminal threshold for harassment.

Important terminology includes content, meaning words, images, audio, video or data; conduct, meaning the broader behaviour surrounding the content; victim or survivor, referring to the person harmed; and offender or suspect, referring to the alleged perpetrator. In professional writing, “alleged” should be used until responsibility is established through an appropriate process.

2. Cyberstalking and harassment

Cyberstalking

Cyberstalking is a pattern of unwanted digital monitoring, contact or surveillance that causes fear, distress or a reasonable concern for safety. It may involve repeated messages, tracking through accounts or devices, impersonation, monitoring social-media activity, contacting friends and employers, publishing personal information, or using spyware. The defining feature is usually persistence combined with intrusion and a harmful effect or risk.

Cyberstalking can be committed by a stranger, acquaintance, former partner or person in a position of authority. Digital tools may extend intimate-partner abuse beyond the physical home. Location sharing, shared passwords and connected devices can become instruments of coercive control.

Harassment

Online harassment generally involves unwanted conduct that causes distress, humiliation, alarm or substantial interference with a person’s life. Depending on the law, it may require repetition, a discriminatory motive, a threat, or proof that the accused knew or should have known the conduct would cause harm. Examples include coordinated abuse, degrading messages, sexualised comments, persistent unwanted contact and targeted insults.

Not every offensive statement is criminal harassment. Legal systems commonly protect robust debate, satire and unpopular opinions. Relevant distinctions include:

• criticism of an idea versus targeting a person with intimidation;
• a single unpleasant remark versus sustained conduct;
• public-interest reporting versus disclosure intended to endanger;
• expression of opinion versus a factual allegation presented as truth;
• lawful protest versus threats or coercive interference.

Evidence and professional response

Victims should, where safe, preserve original messages, URLs, usernames, timestamps and platform notices. Screenshots are useful but may omit metadata or context. A trusted person, advocate, institution or law-enforcement body can help with preservation. Victims should not be required to confront the alleged offender. Investigators should document chronology, authenticate material where possible, examine account ownership and avoid unnecessarily reproducing harmful content.

3. Threats and intimidation

A threat is a communication expressing an intention to cause harm or indicating that harm will follow unless a demand is met. Threat offences vary considerably. Some laws focus on the sender’s intention; others ask whether a reasonable person would perceive the message as a serious threat. Context matters: words, accompanying images, prior conduct, access to weapons, timing, relationship and the recipient’s circumstances may all be relevant.

Threats may concern physical injury, sexual violence, property damage, disclosure of private information or harm to another person. Extortion occurs when a demand is backed by threatened harm or exposure. An online message such as “pay me or I will release your private images” may therefore involve both a threat and an extortionate demand.

A threat should not be dismissed merely because it was sent anonymously, from a temporary account or as an apparent joke. Conversely, investigators must avoid treating every angry or metaphorical statement as a criminal threat without examining context and legal requirements. Immediate danger should be reported to emergency services. Safety planning may include changing account security, informing trusted contacts, reviewing physical routines and seeking specialist support.

4. Cyberbullying

Cyberbullying is repeated or severe digital bullying involving humiliation, exclusion, ridicule, abuse or social aggression. It commonly occurs among children and young people but can affect adults in workplaces, professional groups and public communities. Digital bullying can be especially harmful because content may be persistent, searchable, replicable and visible to a large audience.

The classic bullying relationship includes a power imbalance. Online, power may arise from popularity, technical skill, access to private material, social status, anonymity or the ability to mobilise a group. A single act—such as releasing an intimate image—may produce continuing harm even if the upload occurred only once.

Schools and employers should use clear, proportionate procedures. They should distinguish bullying from ordinary disagreement, protect the targeted person from retaliation, preserve confidentiality, investigate fairly and provide appeal routes. Disciplinary responses should be educational and proportionate, while serious threats, exploitation or abuse should be referred to appropriate authorities.

5. Consent, privacy and image-based abuse

Consent is central to the lawful and ethical use of intimate material. Consent must be informed, voluntary, specific and capable of withdrawal. Consent to take an image is not automatically consent to send it to another person. Consent to send it privately is not consent to post it publicly. Consent to one recipient is not consent to a group or platform.

Consent cannot normally be valid where it is obtained through force, deception about the nature of the act, incapacity, abuse of authority or exploitation. The age of the person is critical: a minor cannot be treated as an adult participant in sexual-image production merely because they appeared willing. Local law determines precise age thresholds and offences.

Image-based abuse

Image-based abuse is the non-consensual creation, acquisition, publication, distribution or threat of distribution of intimate or sexual material. It is often called “revenge pornography,” but that label can be misleading: motives include control, humiliation, financial gain, entertainment, coercion and sexual exploitation. The harm may include loss of employment, social isolation, physical danger and continuing re-circulation.

Potential offences include unlawful disclosure of private images, voyeurism, extortion, harassment, identity misuse and data-protection violations. Civil remedies may include injunctions, damages and takedown orders. Platform policies may provide additional reporting mechanisms, but platform removal does not erase criminal responsibility.

Victims should avoid negotiating with an extortionist where doing so increases risk. They should preserve demands and account information, report content through the platform and contact appropriate specialist services or police. Material involving a child should never be downloaded, forwarded or retained unnecessarily; it should be reported through authorised channels.

6. Sextortion

Sextortion is sexual coercion involving a demand—often for money, more images, sexual acts or continued contact—backed by threatened disclosure or other harm. The perpetrator may obtain material through deception, hacking, grooming, a dating relationship or a consensual exchange. The existence of an original consensual image does not justify coercive control or publication.

A victim is not responsible for the offender’s decision to threaten or distribute material. Common responses such as paying, sending further images or remaining online rarely guarantee safety and may provide additional leverage. A practical response is to stop communication if safe, preserve evidence, secure accounts, report the account, notify a trusted person and seek urgent help where a child or immediate physical danger is involved.

Professionals should use non-judgmental language. Blaming a victim can suppress reporting and impair investigation. Institutions should also consider the risk of self-harm, family violence, homelessness or retaliation when planning intervention.

7. Child sexual abuse material

Child sexual abuse material (CSAM) is material that depicts, describes or facilitates the sexual abuse or sexual exploitation of children, as defined by applicable law. The term is preferred to expressions that imply consent or treat abuse as entertainment. Production, possession, access, distribution, solicitation and knowingly making such material available are serious offences in many jurisdictions. Laws may also cover computer-generated, manipulated or pseudo-photographic material, but definitions vary.

The primary issue is protection of the child and prevention of further circulation. A person who encounters suspected CSAM should not search for more, download it, share it, email it to investigators or post warnings containing the material. The correct action is to record only the minimum safe information needed for reporting—such as the account, URL or platform—and use a recognised reporting authority, child-protection body or law-enforcement channel. Emergency services should be contacted if a child faces immediate danger.

Grooming is the deliberate building of trust or emotional dependence to facilitate sexual exploitation. It may include gifts, secrecy, sexualised conversation, movement to private platforms, threats and gradual boundary testing. Grooming can occur without an in-person meeting. Moderators, educators and investigators should recognise behavioural patterns without assuming that every friendly interaction is criminal; evidence and legal thresholds remain important.

8. Hate-related offences and unlawful publication

Hate-related conduct targets people because of characteristics such as race, ethnicity, religion, nationality, sex, sexual orientation, gender identity or disability. It may take the form of threats, incitement to violence, discriminatory harassment, targeted dehumanisation or the unlawful dissemination of prohibited material. Some legal systems criminalise hate speech broadly; others criminalise only threats, incitement or conduct connected to another offence and protect offensive expression more strongly.

Three distinctions are essential:

1. Prejudice is not always a complete offence. A hateful belief may be morally objectionable but not criminal unless expressed or acted upon in a prohibited way.
2. Offence is not identical to incitement. Offensive criticism may be protected, while intentional encouragement of imminent violence may not be.
3. Context is decisive. Audience, medium, reach, intent, likelihood of harm, historical vulnerability and whether the statement is factual or rhetorical may matter.

Unlawful publication may include publishing private information without lawful basis, defamatory falsehoods, confidential records, court-prohibited material, material identifying protected victims, or sexual images without consent. Publication can occur through websites, social networks, messaging groups, livestreams and search-engine indexing.

The right to privacy does not make all embarrassing information secret, and freedom of expression does not create a general right to publish another person’s intimate data. Professional decision-makers should apply necessity, proportionality, accuracy, public interest and the least intrusive means. Takedown requests should identify the precise material, the legal or policy basis, the harm and the requested remedy.

9. Speech rights, platform governance and due process

Cybercrime law operates within constitutional, human-rights and administrative principles. Restrictions on expression should normally be lawful, legitimate, necessary and proportionate. Vague rules can chill legitimate journalism, whistleblowing, satire, academic research and political dissent. Over-removal can also conceal evidence and silence victims.

Platforms are not neutral in their effects. Recommendation systems can amplify harassment, while automated moderation may misunderstand reclaimed language, journalism or minority dialects. Artificial intelligence can help detect duplicate abuse images, identify coordinated campaigns and prioritise urgent reports, but it may produce false positives, miss coded threats and reproduce bias. Automated decisions should therefore include human review, explanations, secure escalation and appeal mechanisms.

A sound governance model includes:

• clear definitions and accessible rules;
• risk-based moderation;
• preservation procedures for lawful investigations;
• privacy and data-minimisation safeguards;
• special protection for children;
• transparent user notification and appeal;
• independent oversight and regular impact assessment.

Platforms should not expose victims’ identities unnecessarily or require them to repeatedly upload traumatic material. Investigators should seek the least amount of personal data necessary and follow lawful authorisation procedures.

10. Victim protection and evidence-informed practice

A trauma-informed response recognises that victims may experience fear, shame, anger, confusion and loss of control. First responders should listen, avoid blame, explain options and respect the victim’s decisions where law permits. Safety is more important than perfect evidence.

A practical response pathway is:

1. Assess immediate danger. Consider threats, stalking, access to the victim, child safety and self-harm risk.
2. Stabilise accounts and devices. Change passwords using a safe device, enable multi-factor authentication, review active sessions and check location sharing.
3. Preserve evidence safely. Keep messages, URLs, dates, account identifiers and platform responses; do not redistribute illegal material.
4. Report through suitable channels. Use emergency services, police, child-protection agencies, platform reporting and specialist support as appropriate.
5. Reduce ongoing exposure. Block or restrict accounts when safe, adjust privacy settings and ask trusted contacts not to forward content.
6. Provide continuing support. Include counselling, legal assistance, medical care, workplace or school protection and practical safety planning.

Evidence should be collected lawfully. Digital records may be altered, deleted or generated by impersonators. Investigators should preserve original sources, record acquisition methods, maintain chain of custody and respect the distinction between suspicion and proof.

Illustrative case study: “The private group”

Background

A fictional university student, Amina, sends a consensual intimate image to her former partner, Leo. After their relationship ends, Leo posts the image in a private messaging group and threatens to send it to Amina’s classmates unless she resumes the relationship. Members of the group forward it to other accounts. One participant adds racist insults about Amina’s ethnic background.

Analysis

The initial exchange does not establish consent to redistribution. Leo’s demand and threatened disclosure may constitute sextortion, harassment, extortion or image-based abuse under applicable law. The forwarding participants may incur liability if they knowingly distribute prohibited material. The racist abuse may support a hate-related offence or an aggravated penalty, depending on local law. The university’s disciplinary code may apply independently of criminal proceedings.

Appropriate approach

Amina should be offered immediate safety assessment, confidential support and help preserving messages, account identifiers, URLs and threats. She should not be asked to circulate the image for proof. The university should restrict further exposure, protect her from retaliation and refer suspected criminal conduct to the appropriate authority with her informed participation where possible. The platform should remove copies, preserve relevant records under lawful process and prevent re-upload where technically feasible.

Outcome and learning points

The case illustrates that consent is specific, that digital redistribution multiplies harm, and that several legal categories can overlap. It also demonstrates why victim protection, evidence preservation, platform action and due process must operate together. A racist insult should not obscure the privacy violation, and the privacy violation should not prevent careful analysis of speech rights and evidential requirements.

Challenges and limitations

Cyber-offences are difficult to investigate because perpetrators may use pseudonyms, foreign services, encryption, disappearing messages and manipulated media. Cross-border evidence collection can be slow. Victims may under-report because of stigma or fear of retaliation. Laws may be technologically outdated, inconsistently enforced or overly broad. Automated moderation can remove lawful expression while missing coded abuse. Finally, criminal prosecution cannot by itself repair reputational, psychological or relational harm.

Effective policy therefore combines criminal law with civil remedies, education, platform accountability, safeguarding, digital literacy and accessible support services. Prevention should address gender-based violence, racism, unequal power and the social conditions that enable abuse rather than relying only on technical controls.

Best-practice principles

• Treat consent as specific, informed, voluntary and revocable.
• Separate criticism and political expression from threats, coercion and targeted abuse.
• Use precise, non-stigmatising language, especially concerning children and survivors.
• Preserve evidence without copying or redistributing illegal material.
• Prioritise immediate safety and confidentiality.
• Apply proportionate, transparent and reviewable moderation.
• Protect children through safeguarding, reporting and specialist intervention.
• Avoid victim blaming and provide accessible remedies.
• Verify jurisdiction and statutory definitions before making legal conclusions.
• Combine technical, legal, institutional and psychosocial responses.

Chapter Summary

Harmful online content and interpersonal offences cover a wide range of conduct, including cyberstalking, harassment, threats, cyberbullying, image-based abuse, sextortion, CSAM, hate-related offences and unlawful publication. Classification depends on conduct, consent, intent, impact, context and jurisdiction. Consent to create or privately send material is not consent to publish it. Threats and sextortion require urgent, safety-focused responses, while suspected CSAM must be reported through authorised channels without downloading or forwarding it. Hate-related regulation must distinguish serious threats and incitement from protected expression. Victim-centred practice requires safety assessment, careful evidence preservation, privacy, support and due process. Platforms and institutions should use transparent, proportionate and reviewable governance, including careful human oversight of AI moderation.

Key Terms

Cyberstalking: Repeated digital monitoring or contact that causes fear, distress or safety concerns.

Harassment: Unwanted conduct that causes distress, alarm, humiliation or substantial interference.

Threat: A communication indicating intended or likely harm.

Cyberbullying: Digital bullying involving repeated or severe humiliation, exclusion or abuse.

Consent: Informed, voluntary, specific and revocable agreement.

Image-based abuse: Non-consensual creation, acquisition, disclosure or distribution of intimate material.

Sextortion: Sexual coercion backed by threatened disclosure or other harm.

Child sexual abuse material (CSAM): Material depicting or facilitating sexual abuse or exploitation of a child.

Grooming: Building trust or dependence to facilitate sexual exploitation.

Hate-related offence: Prohibited conduct directed at a protected group or characteristic.

Incitement: Encouragement or provocation of unlawful or harmful action.

Unlawful publication: Dissemination prohibited by law, such as private, confidential or illegal material.

Doxxing: Publishing personal information to expose, intimidate or endanger a person.

Extortion: Obtaining something through threats or coercion.

Impersonation: Pretending to be another person online, often to deceive or harm.

Digital evidence: Electronically stored information relevant to an investigation or proceeding.

Chain of custody: Documented control and handling of evidence.

Trauma-informed practice: Support that recognises harm and avoids blame or re-traumatisation.

Content moderation: Processes used to assess, restrict or remove digital content.

Due process: Fair procedures, notice, impartial assessment and opportunity to respond.

Review Questions

5 Short Answer Questions

1. What distinguishes cyberstalking from an isolated offensive message?
2. Why does consent to create an intimate image not necessarily authorise its publication?
3. Define sextortion and identify two common demands made by offenders.
4. Why should suspected CSAM not be downloaded or forwarded?
5. State two factors that help distinguish protected expression from an unlawful threat.

5 Long Answer Questions

1. Explain the legal and practical differences among harassment, cyberbullying, cyberstalking and threats.
2. Discuss consent, privacy and image-based abuse using relevant examples.
3. Explain the risks created by grooming and sextortion and propose a victim-centred response.
4. Analyse the tension between hate-speech regulation and freedom of expression.
5. Describe how platforms, universities and investigators should cooperate while preserving due process.

3 Analytical / Application Questions

1. A former partner threatens to publish private images unless the victim meets them. Analyse the possible offences, evidence and immediate safety measures.
2. A student posts an offensive political opinion, and others demand criminal prosecution. Apply the chapter’s framework to determine what further facts are needed.
3. A platform’s AI system removes a journalist’s report containing abusive material but leaves coded threats online. Evaluate the governance failures and propose safeguards.

Multiple Choice Questions with Answers

Question 1

Which feature most strongly distinguishes cyberstalking from a single offensive comment?

A. The message uses a social-media platform

B. The conduct forms a persistent pattern that causes fear or safety concerns

C. The sender uses a pseudonym

D. The recipient disagrees with the sender

Correct Answer: B. The conduct forms a persistent pattern that causes fear or safety concerns

Question 2

Consent to send an intimate image privately generally means that the recipient may:

A. Publish it publicly

B. Forward it to friends

C. Use it only within the scope of the consent given

D. Threaten disclosure if the relationship ends

Correct Answer: C. Use it only within the scope of the consent given

Question 3

What is sextortion?

A. A lawful age-verification process

B. Sexual coercion involving threatened disclosure or other harm

C. A consensual exchange of intimate images

D. A platform’s removal of sexual content

Correct Answer: B. Sexual coercion involving threatened disclosure or other harm

Question 4

What is the safest general response to suspected child sexual abuse material?

A. Download it to verify its authenticity

B. Forward it to a group for identification

C. Post a warning containing the material

D. Report it through an authorised channel without redistributing it

Correct Answer: D. Report it through an authorised channel without redistributing it

Question 5

Which factor is most relevant when assessing whether an online statement is a serious threat?

A. The colour of the website

B. The sender’s typing speed

C. The wording, context, prior conduct and perceived risk

D. Whether the recipient likes the sender

Correct Answer: C. The wording, context, prior conduct and perceived risk

Question 6

Which statement about consent is correct?

A. Consent is automatically permanent

B. Consent to one recipient is consent to all recipients

C. Consent may be specific and capable of withdrawal

D. Consent is irrelevant when material is digital

Correct Answer: C. Consent may be specific and capable of withdrawal

Question 7

Which response best reflects trauma-informed practice?

A. Blaming the victim for sharing an image

B. Requiring the victim to confront the alleged offender

C. Listening without blame and prioritising safety and informed choices

D. Publishing the evidence to gain public support

Correct Answer: C. Listening without blame and prioritising safety and informed choices

Question 8

Why can AI moderation require human review?

A. AI cannot process any digital content

B. Automated systems may produce false positives, miss coded abuse or reproduce bias

C. Human review always eliminates all errors

D. AI systems are legally incapable of detecting threats

Correct Answer: B. Automated systems may produce false positives, miss coded abuse or reproduce bias

Question 9

Which distinction is central to analysing hate-related speech?

A. All offensive expression is automatically criminal

B. All political expression is automatically protected

C. Threats or incitement may be treated differently from offensive opinion

D. Only anonymous speech can be hateful

Correct Answer: C. Threats or incitement may be treated differently from offensive opinion

Question 10

What is chain of custody?

A. A method for increasing social-media followers

B. A record of evidence control and handling

C. A platform’s advertising policy

D. A technique for creating anonymous accounts

Correct Answer: B. A record of evidence control and handling