zoltner

AI Child Safety News / archive

September 2026: 116 stories on AI and child safety

Every story recorded for September 2026 that is in this archive, newest first. Each entry gives a short summary and links to the original report. Read the September 2026 summary

Search all stories

  1. UK culture secretary says children’s social media ban is only a staging post in online safety drive

    · The Guardian

    UK Culture Secretary Lisa Nandy told The Guardian that the planned under-16 social media ban was one step in a wider effort to address online harms. The paper reported that the ban was due to take effect by March 2027 and that further restrictions for 16- and 17-year-olds had no set date.

    Legislation and rulemaking; Age limits and social media bans · social media ban, United Kingdom, Lisa Nandy

  2. Xbox marks longstanding accounts as age-verified adults without an ID check

    · digitalcitizen.life

    On October 4, Digital Citizen reported that Xbox players received notices saying their accounts were estimated to be adult based on longstanding activity, with no ID check. Microsoft has not published the criteria, the number of accounts switched or whether the status can be reversed.

    Age assurance and verification; Gaming and virtual worlds · Xbox, gaming platforms, age assurance, Microsoft

  3. Iowa attorney general settles child-safety suit with Meta for about $126M

    · openclassactions.org

    Iowa is guaranteed about $126.5M over ten years under Meta’s multistate deal, worth up to $17.1B to 47 states. The settlement also creates a $25M restitution fund, a two-hour daily limit for teens and stronger age verification. The allegations are unproven.

    Lawsuits and settlements; Age assurance and verification · Meta, settlement, Iowa (US), age assurance, screen time limits

  4. India tells Supreme Court it will bar under-18s from creating social media accounts

    · Tech Times

    On 28 September the government told India’s Supreme Court it will amend the IT Intermediary Rules to bar platforms from letting under-18s create accounts. The push rests on the Indian Contract Act, under which a minor’s agreements are void. The case has been adjourned.

    Age limits and social media bans; Legislation and rulemaking; Lawsuits and settlements · India, Supreme Court

  5. Bureau of Investigative Journalism reports police failed to charge anyone over Grok deepfake requestsAllegation

    · thebureauinvestigates.com

    Police in England and Wales have been unable to find and charge people who asked X’s Grok chatbot for sexual deepfakes, and xAI did not answer police requests, the Bureau of Investigative Journalism reported on September 29. Two women have taken civil action against xAI. The allegations are unproven.

    AI-generated abuse imagery; AI chatbots and companions; Criminal cases and prosecutions · Grok, criminal prosecution, deepfakes, xAI, X (Twitter)

  6. Roblox settles with Kansas for more than $10M and adds child-safety measures

    · kctv5.com

    Roblox agreed to more than $10 million in commitments to settle with Kansas, including $9.8 million to the state over three years. The agreement requires age-assurance checks, linked parent accounts, default limits on chat with unknown adults for under-18s and stronger moderation. The allegations are unproven.

    Lawsuits and settlements; Gaming and virtual worlds; Legislation and rulemaking · Roblox, gaming platforms, settlement, Kansas (US), content moderation

  7. UK confirms under-16 social media ban to start March 2027

    · cybernews.com

    The Culture Secretary set the start date, with a possible exemption for platforms judged safe enough. Ofcom’s age-assurance assessment is due by the end of October.

    Age limits and social media bans; Legislation and rulemaking · social media ban, United Kingdom, Ofcom

  8. Discord’s second age-verification rollout hit by bugs and backlash

    · kotaku.com

    The relaunched system sorted some users into the wrong age group and left others unsorted. Discord added verification options and vendor transparency and says more than 90% of users will not need to confirm age.

    Age assurance and verification; Gaming and virtual worlds · Discord, gaming platforms, age assurance

  9. Florida attorney general seeks temporary injunction to halt OpenAI model development

    · Axios

    Florida Attorney General James Uthmeier asked a court on September 28, 2026 for a temporary injunction against OpenAI and ChatGPT, saying the company cannot properly regulate its own technology. The motion alleges ChatGPT violates the Children’s Online Privacy Protection Act by collecting data from under-13s without parental consent. The allegations are unproven.

    Lawsuits and settlements; AI chatbots and companions · OpenAI, injunction, Florida (US), ChatGPT, James Uthmeier

  10. South Korean undercover digital sex crime investigations more than tripled this yearAllegation

    · The Korea Times

    Korean police ran 598 undercover investigations into digital sex crimes from January to August, up from 179 a year earlier, as deepfake offenses grew. Undercover work involving conversations with children and adolescents for sexual exploitation led to 338 suspects apprehended and six formally arrested. The allegations are unproven.

    Lawsuits and settlements; Regulatory enforcement and investigations · South Korea, criminal prosecution, deepfakes

  11. Bihar police book a social media user over purported AI videos of a harassed teenagerAllegation

    · Scroll.in

    Bihar Police booked a social media user for uploading purportedly AI-generated videos of a minor girl molested in Jamui district on 19 September. Police said the videos were misleading and fake and revealed her identity. The case cites the POCSO Act, the IT Act, the Bharatiya Nyaya Sanhita and the Juvenile Justice Act. The allegations are unproven.

    Criminal cases and prosecutions; Lawsuits and settlements · criminal prosecution, AI-enabled harassment

  12. Delhi report urged a national policy and platform duty of care for children’s digital use

    · EducationWorld

    The Centre for Legal Action and Behaviour Change report, launched at a UN General Assembly side event, compared age limits in more than 40 countries and school phone rules in 110. It proposed banning accounts for under-15s, an independent online safety regulator and penalties on platforms, not children or parents.

    Legislation and rulemaking · Digital Duty of Care (Australia), UN, age limits, international cooperation, school phone bans

  13. Study links heavier chatbot use for friendship to later loneliness among middle schoolersPreprint

    · The Wall Street Journal

    The Wall Street Journal reported that a University of North Carolina study of more than 2,300 sixth through eighth graders at 22 schools found that students who used AI more often for friendship were more likely to report greater loneliness later. The paper was under peer review at the time.

    Research and evidence; Education and schools; Mental health and self-harm · mental health, chatbots, North Carolina (US)

  14. Survey of 141 researchers set out guardrails for adolescent AI chatbot use

    · AI & Society (Springer Nature)

    A mixed-methods study in AI & Society surveyed 141 researchers and interviewed 15. Participants prioritized severe harms, AI as a relational actor and overreliance, and named three immediate guardrails: enforceable youth-centered regulation, developmentally informed design and multidimensional AI literacy.

    Research and evidence; AI chatbots and companions; Legislation and rulemaking · survey, chatbots, AI literacy

  15. Safe AI for Children Alliance urges children’s rights be central to frontier AI governance

    · Safe AI for Children Alliance

    The Safe AI for Children Alliance published a statement on 25 September backing an international Call for Control of Frontier AI Models. It urged that children’s rights, safety and long-term interests be central to its proposals, and supported mandatory pre-deployment testing, independent evaluation and coordinated government oversight.

    Legislation and rulemaking · AI governance, children’s rights

  16. UN rights council committee presents study on technology-facilitated gender violence

    · transcripts.un.org

    The UN Human Rights Council Advisory Committee presented its study on technology-facilitated gender-based violence affecting women and girls at the council’s 63rd session on 24 September. Delegations including the EU, Brazil and France responded.

    Research and evidence · UN, international cooperation, Brazil, European Union, France

  17. Philippines lifts Discord ban after child-safety commitments

    · pna.gov.ph

    The Philippines’ Department of Information and Communications Technology said on September 24 that the government had lifted its restriction on Discord after the company pledged to strengthen child safety measures. Officials said the commitments addressed grooming and extremist content and that they would work with Discord to set timelines.

    Legislation and rulemaking; Gaming and virtual worlds; Industry policy and partnerships · Discord, gaming platforms, Philippines, grooming, violent extremism

  18. TikTok to pay Alabama at least $100M and adopt teen limits in settlement

    · alabamaag.gov

    On September 25, 2026, Alabama announced a settlement of its suit against TikTok days before trial: at least $100 million, up to $300 million if conditions are met. TikTok must adopt a two-hour daily limit for teens, midnight-to-6am access restrictions, a teen ban on cosmetic filters and stronger parental controls. The allegations are unproven.

    Lawsuits and settlements; Legislation and rulemaking; Platform safety features and design · TikTok, settlement, Alabama (US), parental controls, screen time limits

  19. Indonesia’s minister urged stronger child protection in the AI era at the UN

    · RRI

    Minister Meutya Hafid told a high-level forum on the sidelines of the UN General Assembly that about 70 million Indonesian children are growing up in an AI-shaped digital ecosystem. She said governments, platforms, industry, parents and schools each have a role.

    Legislation and rulemaking · UN, international cooperation, Indonesia

  20. University of East London: women and girls disproportionately targeted in online gaming

    · Phys.org

    Early findings from a University of East London project drew on 10 focus groups with 46 young people aged 12 to 19, a literature review and expert interviews. Women and girls were disproportionately targeted by antisocial behavior. A nationally representative survey is planned.

    Research and evidence; Education and schools; Gaming and virtual worlds · survey

  21. London court jails man for using AI to turn photos of children into explicit images

    · Metropolitan Police

    The Metropolitan Police said a man was sentenced on 24 September to two years and four months at Wood Green Crown Court for using AI to turn photographs of women and children into explicit images. Officers found about 1,800 indecent images of children on his devices.

    AI-generated abuse imagery; Lawsuits and settlements · criminal prosecution

  22. Judge lets Los Angeles County child-exploitation suit against Roblox proceedAllegation

    · iclg.com

    A Los Angeles County Superior Court judge allowed the county’s child-exploitation suit against Roblox to proceed on September 23. ICLG reported that the judge rejected Roblox’s Section 230 and First Amendment arguments at the dismissal stage because the claims concerned the company’s own age-verification practices. The allegations are unproven.

    Lawsuits and settlements; Age assurance and verification; Grooming, sextortion and exploitation · Roblox, gaming platforms, First Amendment, Section 230, age assurance

  23. TikTok drops appeals and accepts £12.7 million UK ICO fine over under-13 dataAllegation

    · Information Commissioner’s Office

    TikTok withdrew its appeal of the ICO’s £12.7 million penalty for mishandling children’s personal data, making the penalty final. It also withdrew a separate appeal against an information notice, allowing the ICO to resume investigating how TikTok processes 13- to 17-year-olds’ personal data in its recommendation systems.

    Regulatory enforcement and investigations; Lawsuits and settlements; Age limits and social media bans · TikTok, ICO, United Kingdom, privacy and data protection, algorithmic feeds

  24. California federal court denies TikTok’s motion to dismiss minors’ COPPA class claims

    · blogs.duanemorris.com

    On September 24, 2026, a federal judge in California denied TikTok’s motion to dismiss minors’ COPPA class claims. The court held the pleadings did not show the named plaintiffs belonged to two earlier settlement classes, worth over $93 million combined, that TikTok said barred the claims. The allegations are unproven.

    Lawsuits and settlements; Privacy and children’s data · TikTok, COPPA, California (US), United States, settlement

  25. Vals AI found critical safety failures in 178 of 648 simulated teen chatbot chats

    · Vals AI

    Vals AI ran 648 simulated conversations across 72 clinician-written scenarios and nine models. At least one critical safety check failed in 178 conversations, and 111 of those failures came later in the exchange, a median of five replies after the concern first appeared.

    Research and evidence; AI chatbots and companions · chatbots

  26. Ofcom opens investigation into Pornhub operator Aylo over age checksAllegation

    · Ofcom

    On September 23, Ofcom opened an Online Safety Act investigation into Aylo, operator of Pornhub, over concern that it did not do sufficient due diligence and testing before adopting an age assurance process that relies on signals from Apple. The allegations are unproven.

    Regulatory enforcement and investigations; Age assurance and verification · Pornhub / Aylo, Ofcom, age assurance, pornography, Apple

  27. YouTube CEO dismisses joining Meta’s $18 billion child-safety settlement

    · cnn.com

    On September 23, YouTube CEO Neal Mohan told CNN YouTube already has safety measures and dismissed joining Meta’s multistate settlement. Meta said it would pay only 70% unless TikTok and YouTube each pay about $6 billion and adopt similar changes. California’s attorney general said states are willing to sue.

    Lawsuits and settlements · YouTube, Meta, settlement, TikTok, Neal Mohan

  28. Von der Leyen and Tim Cook discuss EU Kids Act app-store age checks

    · 2eu.brussels

    The two met at the UN General Assembly. The EU proposal would require app stores to age-rate every app and gate minors’ access. No Apple commitment was announced.

    Legislation and rulemaking; Age assurance and verification · EU Kids Act, app stores, European Union, Ursula von der Leyen, age assurance

  29. Philippine regulators act against Meta, Roblox, Reddit and Discord over child safetyAllegation

    · sunstar.com.ph

    The Cybercrime Investigation and Coordinating Center gave Reddit and Discord 24 hours to appoint local representatives after a third school shooting in three months, warning of possible restrictions. The privacy commission issued show cause orders to Meta, Roblox, Reddit and Discord over data privacy registration non-compliance. The allegations are unproven.

    Gaming and virtual worlds; Regulatory enforcement and investigations; Lawsuits and settlements · gaming platforms, Discord, Reddit, Meta, Roblox

  30. Philippine privacy regulator issues show-cause orders to Meta, Roblox, Reddit and DiscordAllegation

    · manilatimes.net

    The Philippine National Privacy Commission issued show-cause orders to Meta, Roblox, Reddit and Discord on September 22 over data-privacy registration requirements, as part of broader oversight aimed at protecting children’s personal data. The allegations are unproven.

    Gaming and virtual worlds; Regulatory enforcement and investigations; Privacy and children’s data · gaming platforms, privacy and data protection, Discord, Meta, Reddit

  31. Discord launches global privacy-preserving age assurance with over 90% auto-classified

    · discord.com

    Discord announced on September 22 that it would begin assigning users to age groups using account signals, with non-biometric options for people asked to confirm their age. The company said it would restrict teens’ access to age-gated spaces and route messages from non-friends to a requests inbox. The account-signal model was unavailable in the UK and Australia.

    Age assurance and verification; Gaming and virtual worlds; Industry policy and partnerships · Discord, gaming platforms, age assurance, privacy and data protection, biometric data

  32. British Columbia sues OpenAI and Sam Altman over the Tumbler Ridge school shootingAllegation

    · The Guardian

    British Columbia sued OpenAI and Sam Altman in federal court in San Francisco on September 21, alleging that company leaders overruled a recommendation to alert police after ChatGPT conversations about gun violence were flagged in June 2025. The province sought damages for recovery costs and a court order changing how OpenAI handled conversations that could lead to violence. The allegations are unproven.

    Lawsuits and settlements; AI chatbots and companions; Education and schools · OpenAI, Sam Altman, Canada, United Kingdom, school shooting

  33. Google rolls out AI age estimation to Canadians using search and YouTube history

    · dexerto.com

    Google began gradually rolling out a machine-learning age-estimation system in Canada in September 2026, using account signals such as searches and YouTube viewing to infer whether users are under 18. Accounts flagged as minors get default protections, and adults misclassified may need to upload a government ID or take a selfie.

    Age assurance and verification; Platform safety features and design · Google, YouTube, age assurance, Canada, biometric age estimation

  34. Philippine Senate unanimously passes Sagip Bata Act on third reading

    · Philippine Daily Inquirer

    The Philippine Senate approved Senate Bill 1819 on third reading on September 21. It would criminalize grooming and luring of children, including AI-enabled sexualization, require platforms to prevent or disrupt abuse, and replace Republic Act 11930. Penalties range from fines of P100,000 to P10 million and up to 40 years in prison.

    Legislation and rulemaking · United States, Sagip Bata Act (Philippines), Philippines, grooming

  35. Anthony Albanese and Tim Cook: Inside the prime minister meeting at Apple Park to discuss child safety and

    · theage.com.au

    Australian Prime Minister Anthony Albanese met Apple chief executive Tim Cook at Apple Park to discuss child online safety and promote Australia’s under-16 social media restrictions. The Age reported that Apple presented controls allowing parents to limit app use and restrict who could contact their children.

    Legislation and rulemaking · Apple, Anthony Albanese, Australia

  36. The rogue AI Sherlock apps used to humiliate Greater Manchester teens

    · manchestereveningnews.co.uk

    A Greater Manchester youth worker told the Manchester Evening News that some people used AI face-search apps to identify teenagers and humiliate them in videos shared online. He said some young people were also turning to chatbots as confidants. A face-search app developer said its service was intended for adults and prohibited searches involving children.

    Legislation and rulemaking · AI-enabled harassment, chatbots

  37. Hong Kong to study legislative and administrative measures on minors’ use of AI and social media

    · news.gov.hk

    On September 18, Secretary for Justice Paul Lam said Hong Kong would take a multi-pronged approach to protecting minors using AI and social media. The Policy Address set up a Department of Justice-led task force to review risks and consider legislative and administrative measures, and the Education Bureau will introduce education policy. The public will be consulted first.

    Legislation and rulemaking; Research and evidence · Hong Kong, US Department of Justice

  38. Federal judge signals rejection of part of TikTok’s $400M COPPA settlementAllegation

    · CNBC

    A federal judge said he was inclined to reject a request to end a 2019 Federal Trade Commission consent decree and scheduled a hearing. Under TikTok and ByteDance’s proposed settlement of children’s privacy allegations, $100 million of the $400 million payment depended on ending that decree. The allegations are unproven.

    Lawsuits and settlements; Legislation and rulemaking; Privacy and children’s data · TikTok, United States, COPPA, settlement, ByteDance

  39. JAMA Pediatrics survey: 13.4% of US teens saw unwanted sexual images online in four weeks

    · JAMA Network

    A nationally representative pilot survey of 472 US adolescents aged 13 to 17 found 13.4% saw unwanted sexual images online, against 2.0% offline, and 9.7% saw online content encouraging self-harm. Harms varied by platform, and 51.5% reported technology-related sleep disruption.

    Research and evidence; AI-generated abuse imagery · survey, United States, self-harm and suicide

  40. China drafts guidelines requiring minor modes and barring virtual intimacy services for minors

    · China.org.cn

    China’s Cyberspace Administration opened a consultation through October 17 on draft rules that would bar virtual intimate relationship services for minors. The draft would require providers to use a minor mode for social networking and for AI services that could affect the cognition of users under 16.

    Legislation and rulemaking · China, legislative proposal

  41. NPR finds tech industry shaped state chatbot safety bills with broad exemptions

    · NPR

    NPR found that many US state chatbot safety bills were drafted with heavy tech industry input and carried exemptions that could cover the most popular chatbots. NPR counted at least 75 lawsuits against AI developers over alleged chatbot harms, many involving children.

    Legislation and rulemaking; AI chatbots and companions; Lawsuits and settlements · chatbots, United States

  42. UK advertising regulator bans first nudify app ads on Meta

    · The Independent

    The UK Advertising Standards Authority banned a series of Meta ads for nudify and AI companion apps, saying some encouraged creating sexually explicit images of women without consent and one sexualized a person who appeared under 18. It was understood to be the ASA’s first such ban. Meta said the ads were removed before being reported.

    AI-generated abuse imagery; Legislation and rulemaking; Regulatory enforcement and investigations · Meta, nudification, United Kingdom, companion chatbots, sexualized AI content

  43. Meta expands Threads parental supervision to Asia-Pacific for under-16s

    · Meta Newsroom

    Family Center adds time spent, daily limits, sleep mode and privacy management for teens in the region.

    Platform safety features and design; Age limits and social media bans · Meta, Threads, parental controls, privacy and data protection, screen time limits

  44. Lawsuit alleges Grok used a Wyoming woman’s childhood photo to generate 7,000 CSAM imagesAllegation

    · cowboystatedaily.com

    Part of a class action naming xAI. The plaintiff alleges her stepfather chose Grok because it was less restrictive, and accuses xAI of obstructing the investigation. The allegations are unproven.

    Lawsuits and settlements; AI-generated abuse imagery; AI chatbots and companions · Grok, AI-generated CSAM, Wyoming (US), xAI, class action

  45. European Commission formally proposes the EU Kids Act

    · iapp.org

    The European Commission unveiled the proposed EU KIDS Act on September 17. It would bar social media accounts for under-13s and allow restricted accounts for 13- and 14-year-olds through a parent or guardian’s account. It also proposed privacy-preserving age assurance and limits on addictive features for designated very large services, including social media, video sharing, games, and AI companions and chatbots.

    Legislation and rulemaking; Age assurance and verification · European Union, EU Kids Act, European Commission, addictive design, age assurance

  46. Snap CEO tells BBC the company is willing to implement default teen time limits

    · digitaltrends.com

    Digital Trends reported on September 17 that Snap CEO Evan Spiegel had told the BBC he was open to a daily Snapchat limit for teens. He offered no commitment or timeline, and the article noted that Snapchat already offered parental oversight through Family Center.

    Platform safety features and design; Legislation and rulemaking · Snap, screen time limits, Evan Spiegel, parental controls

  47. Dutch foundation SOMI files pan-European class action against Snap across 27 EU statesAllegation

    · memesita.com

    On September 16, Dutch foundation SOMI and law firm Spirit Legal filed a collective action against Snapchat covering 27 EU member states. They allege breaches of the Digital Services Act and GDPR, including profiling minors for advertising and use of data from the My AI chatbot, and seek compensation and a platform redesign. The allegations are unproven.

    Legislation and rulemaking; Lawsuits and settlements · Snap, European Union, Netherlands, class action, Digital Services Act

  48. EU to propose a Kids Act restricting under-15s from social media, AI chatbots and games

    · Reuters

    Reuters reported on September 14 that a draft EU KIDS Act would allow 13- and 14-year-olds to use limited social media accounts opened and controlled by parents. The draft would require age checks and measures to avoid addictive designs and harmful feeds. The proposal still needed negotiations with EU countries and the European Parliament.

    Legislation and rulemaking; Platform safety features and design; Age assurance and verification · European Union, chatbots, EU Kids Act, European Parliament, addictive design

  49. France submits new proposal for an under-15 social media ban

    · Le Monde

    On September 14, President Emmanuel Macron said France notified the European Commission of a reworked under-15 social media ban. The Constitutional Council struck down the previous law in mid-August as an unnecessary infringement on freedom of expression.

    Legislation and rulemaking; Age limits and social media bans · France, social media ban, Emmanuel Macron, European Commission, European Union

  50. Canadian Centre and New York Times find X still hosts CSAM; reports say Grok can still make nude imagery

    · socialmediatoday.com

    The Canadian Centre for Child Protection found X still hosts child sexual abuse material, and a New York Times experiment found users still share material authorities had flagged. Reports suggest Grok can still generate nude imagery despite X’s restrictions.

    AI-generated abuse imagery; AI chatbots and companions · X (Twitter), Grok, AI-generated CSAM, Canada, New York (US)

  51. Google announces age-assurance rollout in Canada with default protections for under-18s

    · Google

    Google announced it will gradually roll out machine-learning age estimation in Canada, applying default under-18 protections to users it estimates are minors. Users wrongly estimated as under 18 can correct their age with a government ID or selfie.

    Age assurance and verification; Age limits and social media bans; Industry policy and partnerships · Google, Canada, age assurance, biometric age estimation

  52. Iowa to pay teen Instagram users restitution from Meta settlement

    · desmoinesregister.com

    On September 14, Iowa Attorney General Brenna Bird said $25 million of the state’s Meta settlement share would fund restitution for young Instagram users, which she expects to pay about $100 per Teen Account user. She called it the first such fund in any state. Payment criteria are still being developed.

    Lawsuits and settlements; Platform safety features and design · Instagram, Meta, settlement, Iowa (US), teen accounts

  53. Philippines to lead ASEAN cooperation on online child safety

    · The Philippine Star

    The Philippines’ Cybercrime Investigation and Coordinating Center said it would push for stronger Southeast Asian cooperation against online grooming and other threats to children on social media and gaming platforms. Its director spoke after ScamWatch officials urged use of the country’s 2026 ASEAN chairmanship.

    Legislation and rulemaking · international cooperation, Philippines, grooming

  54. US schools and police warn of viral Cat in the Hat threat trend after teen arrests

    · The Guardian

    US schools and police warned about a viral trend, which began on Snapchat, in which disturbing or AI-generated versions of the Cat in the Hat were used to threaten students, schools and communities. Several teenagers were arrested or charged over posts connected to the trend.

    Criminal cases and prosecutions; Education and schools; Legislation and rulemaking · criminal prosecution, United States, Snap

  55. Australian minister open to softening draft digital duty of care powers

    · ABC News (Australia)

    Australia’s Communications Minister Anika Wells said she was willing to change draft laws that would let her designate additional serious online harms, to win Coalition support. The draft would fine platforms more than A$100 million for failing to reduce foreseeable harm, including child sexual exploitation content and grooming.

    Legislation and rulemaking · Digital Duty of Care (Australia), Australia, Anika Wells, grooming, international cooperation

  56. Police in England and Wales log 16-fold rise in AI-generated child sexual images

    · The Telegraph

    Twenty police forces in England and Wales identified 163 recorded crimes by July 2026 in searches for terms associated with AI-generated sexual images of children, The Telegraph reported. The comparable counts were 10 in 2023, 46 in 2024, and 112 in 2025. Some cases were closed because of evidentiary difficulties.

    AI-generated abuse imagery; Criminal cases and prosecutions · United Kingdom, criminal prosecution, AI-generated CSAM

  57. OpenAI announces research grants on AI and teen development

    · OpenAI

    On September 8, OpenAI committed $5 million to independent research on how generative AI affects the development of teens aged 13 to 17, with individual grants of up to $1 million. Applications were open through October 6, 2026.

    Industry policy and partnerships; AI chatbots and companions; Research and evidence · OpenAI

  58. South Africa says it has no plans to introduce a social media ban for children

    · MyBroadband

    South Africa’s communications minister, Solly Malatsi, said the government had no plans to ban under-16s from social media, citing a lack of conclusive evidence that a blanket ban would keep them off the platforms. The article contrasts that position with draft proposals for age checks on video-sharing platforms.

    Age limits and social media bans; Legislation and rulemaking · social media ban, South Africa, age assurance

  59. Canada’s privacy commissioner finds X and xAI violated privacy law over Grok deepfakes

    · magica.com

    The regulator found no valid consent process for sexualized deepfakes, deemed later safeguards unproven and recommended suspending Grok Imagine. The companies declined but committed to quarterly reports and annual audits.

    Legislation and rulemaking; AI-generated abuse imagery; AI chatbots and companions · Grok, X (Twitter), deepfakes, privacy and data protection, xAI

  60. Young Life study found 35% of surveyed young people have an AI companion

    · Young Life via PR Newswire

    Young Life’s RELATE Project 2026 surveyed 7,105 people aged 13 to 24 in 11 countries. Some 35% reported an AI companion, rising to 45% of 13 to 15 year olds, and 60% of those users said they were more comfortable with it than with some friends or family.

    Research and evidence; AI chatbots and companions · companion chatbots, survey

  61. EU Kids Online survey of 28,465 children found 48% feel safe online

    · London School of Economics

    The EU Kids Online 2026 report surveyed children aged 10 to 16 in 19 countries between April 2025 and April 2026. Only 48% said they feel safe online, 23% had been bothered or upset by something online in the past month, and about two-thirds regularly use AI tools.

    Research and evidence · Safe Online, survey, European Union

  62. Irish regulator report finds children and parents prefer online safety education to a social media ban

    · Coimisiún na Meán

    Coimisiún na Meán’s research report found most children in Ireland think teaching them to stay safe online would be better than banning social media. Almost half of parents and caregivers also favor online safety education.

    Research and evidence; Age limits and social media bans; Education and schools · Ireland, social media ban, Coimisiún na Meán, Safe Online

  63. Microsoft publishes Safe Participation Framework and launches Windows Age API

    · blogs.microsoft.com

    On September 10, Microsoft published a Safe Participation Framework and introduced a Windows Age API that gives developers privacy-preserving age signals. All Copilot users must now sign in, and access is restricted for children under 13. Microsoft also pointed to Xbox family tools and new Windows Family Safety protections.

    Industry policy and partnerships; Platform safety features and design · Microsoft, Copilot, Xbox, age assurance, gaming platforms

  64. California governor signs child online-safety and AI chatbot package into law

    · dig.watch

    On September 10, the bills include SB 1119 on companion-chatbot crisis protocols, parental controls and audits, AB 1709 banning addictive features for under-16s, and expanded CSAM law covering AI-generated imagery of minors.

    Legislation and rulemaking; AI chatbots and companions; AI-generated abuse imagery · chatbots, California (US), California AB 1709, California SB 1119, AI-generated CSAM

  65. Ninth Circuit affirms Roblox waived right to compel arbitration in child-safety class actionAllegation

    · law.com

    The Ninth Circuit upheld the denial of Roblox’s motion to compel arbitration in Uhl v. Roblox. In an unpublished decision, the court held that Roblox waived its arbitration right by taking steps inconsistent with arbitration, including litigating the merits before moving to compel it.

    Lawsuits and settlements; Gaming and virtual worlds · Roblox, arbitration, gaming platforms, class action

  66. TikTok moves to dismiss Florida’s under-14 social-media suit

    · floridaphoenix.com

    TikTok asked a Florida court to dismiss the state’s lawsuit, arguing that HB 3 is unconstitutionally vague and violates the First Amendment. The law bars children under 14 from covered platforms and requires parental consent for 14- and 15-year-olds. Florida alleges TikTok is a public nuisance and misrepresented the content available on its app. The allegations are unproven.

    Legislation and rulemaking; Age limits and social media bans · TikTok, Florida (US), First Amendment, HB 3 (Florida), parental consent

  67. Texas court finds TikTok liable for misleading parents on child-safety protections

    · texasattorneygeneral.gov

    On September 10, Texas Attorney General Ken Paxton announced that District Judge Cory Liu found TikTok liable for misleading parents about its safety standards, saying graphic videos stayed accessible to minors even in Restricted Mode. A trial to determine relief and penalties is expected next month.

    Lawsuits and settlements · TikTok, Texas (US), Ken Paxton

  68. UK to legislate device-level blocks on nude images for under-18s

    · GOV.UK

    The UK government said it would bring forward legislation requiring tech companies to build device-level protections that make it impossible for under-18s to take, view or send nude images. It said talks with Apple and Google fell short of its ambition. Autoplay and personalized feeds would be off by default for 16 and 17 year olds.

    Legislation and rulemaking; Age limits and social media bans; Platform safety features and design · United Kingdom, Apple, Google

  69. New Roblox and Snap grooming lawsuit filed in federal court in CaliforniaAllegation

    · aboutlawsuits.com

    On September 9, a 15-year-old Florida girl alleges inadequate age verification and safety systems at Roblox and Snap allowed a predator to groom and exploit her. The case joins more than 180 consolidated federal cases. The allegations are unproven.

    Lawsuits and settlements; Grooming, sextortion and exploitation; Gaming and virtual worlds · Roblox, Snap, gaming platforms, grooming, United States

  70. Teens increasingly turn to AI chatbots for mental-health support

    · edsource.org

    EdSource reported that a Brown University study published in November 2025 found that about one in eight adolescents had turned to AI chatbots for help with their mental health. A researcher told EdSource that adolescent chatbot use for emotional support had risen by 50% since the study’s publication.

    Research and evidence; AI chatbots and companions; Mental health and self-harm · chatbots, mental health

  71. Xbox hits Australia age-assurance deadline for R18+ downloads

    · tech-insider.org

    On September 9, Under eSafety’s Age-Restricted Material Codes, Microsoft must age-check R18+ purchases and downloads in Australia by Sept 9 using facial estimation, ID or credit card. Games already owned keep working.

    Legislation and rulemaking; Age assurance and verification; Gaming and virtual worlds · Xbox, Australia, gaming platforms, Age-Restricted Material Codes (Australia), Microsoft

  72. India’s NCPCR summons Meta India MD over child-abuse ads on InstagramAllegation

    · timesofindia.indiatimes.com

    On September 9, the National Commission for Protection of Child Rights summoned Meta India’s managing director to examine how child sexual exploitation ads cleared Instagram’s systems. The NHRC issued parallel notices to Indian ministries on Sept 2. The allegations are unproven.

    Lawsuits and settlements; Regulatory enforcement and investigations · India, Instagram, Meta, NCPCR (India), European Commission

  73. Irish media regulator opens investigation into X over age assurance and parental controlsAllegation

    · RTE

    On September 8, Coimisiún na Meán opened a formal investigation into X under the Online Safety Code. The regulator suspects X’s age checks and parental controls may not be effective. The allegations are unproven.

    Regulatory enforcement and investigations; Age assurance and verification; Legislation and rulemaking · X (Twitter), age assurance, parental controls, Ireland, Coimisiún na Meán

  74. Australia releases Digital Duty of Care draft and moves to strengthen under-16 ban enforcement

    · pm.gov.au

    On September 8, the government released a draft Digital Duty of Care bill for public consultation. The prime minister cited penalties of more than $100 million for non-compliant platforms, and the duty extends to games, AI chatbots and private chats. The minister said it covers six child harms, but the transcript names five.

    Age limits and social media bans; Regulatory enforcement and investigations; Lawsuits and settlements · Digital Duty of Care (Australia), Under-16 social media minimum age (Australia), social media ban, Australia, chatbots

  75. TTP report: Meta ran 332 AI-generated CSAM ads, many in India after removal order

    · BBC News

    The BBC reported that the Tech Transparency Project identified 332 Facebook and Instagram ads featuring AI-generated child sexual abuse material across the United States, United Kingdom, European Union, Australia, and India. Of those, 274 ran in August, and 78 appeared in India after officials ordered Meta to remove the material and met with company representatives. Most ads directed users to AI image or video generation apps.

    AI-generated abuse imagery · Meta, India, AI-generated CSAM, Facebook, Instagram

  76. Philippine ministry hopes Roblox national ID age check can be a template for Meta

    · rappler.com

    On September 7, DICT Secretary Henry Aguda told a House budget hearing that Roblox proposed linking its age verification to the Philippine national ID, using API calls that only confirm a user’s age tier. He hoped it could become a template for Meta, which said it would send a technical team to study feasibility.

    Age assurance and verification; Gaming and virtual worlds · Meta, Roblox, age assurance, digital ID, gaming platforms

  77. Apple requires developer social-media declarations for age ratings ahead of iOS 27 Time Allowances

    · webpronews.com

    Apple added social-media capability questions to the App Store Connect age-rating questionnaire to support parental Time Allowances. Beginning in September 2026, developers had to answer them when submitting new apps or updates or seeking notarization for alternative distribution.

    Platform safety features and design; Age assurance and verification · Apple, age ratings, app stores

  78. Australian police report rise in sextortion of young men, with scammers using AI lures

    · ABC News (Australia)

    South Australia Police received reports from 220 male victims of image-based abuse and sextortion in the last financial year, up from 170. The eSafety Commissioner said scammers use AI to generate fake but attractive online characters. An Australian Institute of Criminology report found more than one in 10 adolescents aged 16 to 18 had experienced sexual extortion.

    Research and evidence; Criminal cases and prosecutions; Grooming, sextortion and exploitation · Australia, criminal prosecution, sextortion, eSafety Commissioner

  79. Labor urges Senate to pass stronger enforcement powers for Australia’s under-16 ban

    · SBS News

    Australia’s Labor government urged the Greens and Coalition to pass a bill giving the eSafety Commissioner stronger information-gathering powers and higher fines for breaches of the under-16 social media ban. eSafety found 81.5 percent of 10 to 15 year olds still used an age-restricted platform in March and April.

    Age limits and social media bans; Legislation and rulemaking; Regulatory enforcement and investigations · Under-16 social media minimum age (Australia), social media ban, Australia, United States, eSafety Commissioner

  80. Australia to unveil draft law giving users an opt-out from social media algorithms

    · ABC News (Australia)

    Australia’s government was set to release draft digital duty of care laws requiring platforms to offer users an opt-out from algorithmic feeds and to prevent harm, with fines above $100 million. The eSafety Commissioner and registered researchers would be able to use sock-puppet accounts posing as children to monitor what feeds show young people.

    Legislation and rulemaking; Regulatory enforcement and investigations · algorithmic feeds, Australia, legislative proposal, Digital Duty of Care (Australia), eSafety Commissioner

  81. Philippines and Meta deepen cooperation on child online protection

    · pna.gov.ph

    The Philippines’ Department of Information and Communications Technology reached an agreement with Meta to improve child protection online. The Philippine Information Agency said Meta pledged to detect and remove child exploitation content faster and report it to authorities more quickly.

    Lawsuits and settlements · Meta, Philippines, international cooperation

  82. US court protects private possession of AI-generated CSAM

    · media.ca7.uscourts.gov

    On August 25, the Seventh Circuit affirmed dismissal of a federal possession charge as applied to obscene AI-generated images depicting no actual children that were kept privately at home. The appeal concerned the possession count.

    AI-generated abuse imagery; Lawsuits and settlements · United States, AI-generated CSAM

  83. Utah paused enforcement of its adult-site age verification rule against Pornhub’s owner

    · KSL

    Under a stipulation with Aylo, Utah will not enforce its proposed age verification rule against Aylo sites until a federal judge rules on Aylo’s injunction request or the rule is finalized. Aylo agreed to keep blocking access in Utah. The deal does not cover other companies.

    Age assurance and verification; Regulatory enforcement and investigations; Lawsuits and settlements · Pornhub / Aylo, Utah (US), age assurance, pornography, injunction

  84. UNICEF finds 1.5 million Pakistani children faced tech-facilitated sexual abuse in a year

    · Business Recorder

    UNICEF’s Disrupting Harm in Pakistan report estimated that at least 1.5 million internet-using children aged 12 to 17, or 6.25 percent, faced technology-facilitated sexual exploitation and abuse in one year. More than half told no one, and none reported through formal channels. AI-generated fake images were among the documented risks.

    Research and evidence · UNICEF, Pakistan

  85. Commission designates ChatGPT, Reddit and Roblox under the Digital Services Act

    · European Commission

    The Commission designated ChatGPT as a Very Large Online Search Engine, and Reddit and Roblox as Very Large Online Platforms, which brings the strictest risk assessment and transparency duties.

    Regulatory enforcement and investigations; AI chatbots and companions; Gaming and virtual worlds · ChatGPT, Reddit, Roblox, Digital Services Act, European Commission

  86. Ofcom fines adult site 730,000 pounds under Online Safety Act for missing age checks

    · wired-gov.net

    Ofcom fined the provider of Xgroovy.com 700,000 pounds for lacking highly effective age assurance between July and November 2025, plus 30,000 pounds for ignoring an information request. The site has since introduced age checks and blocked UK access. Ofcom also issued provisional decisions against two more providers, Cyberitic and ZD Media.

    Regulatory enforcement and investigations; Age assurance and verification; Legislation and rulemaking · Ofcom, age assurance, Online Safety Act (UK), pornography, United Kingdom

  87. Federal judge denies xAI’s second bid to block Minnesota’s nudify ban

    · ibtimes.co.uk

    The judge found xAI showed no irreparable harm. Minnesota’s law, with civil penalties up to $500k per violation, stays enforceable during xAI’s First Amendment challenge.

    Lawsuits and settlements; AI-generated abuse imagery; Legislation and rulemaking · xAI, Minnesota (US), nudification, United States, First Amendment

  88. TikTok withdraws from US House roundtable

    · chinaselectcommittee.house.gov

    On September 3, Representative John Moolenaar, chair of the House Select Committee on China, said TikTok backed out of appearing at the committee’s September roundtable on its algorithm and China ties. TikTok was to send its Chief Security Officer. He said TikTok cited a wish to avoid broader scrutiny of its child safety practices.

    Legislation and rulemaking · TikTok, United States, China, algorithmic feeds

  89. UNICEF study finds 1 in 5 children in 21 countries faced tech-facilitated sexual abuse

    · unicef.org

    UNICEF reported on September 3 that an estimated 20 million children aged 12 to 17, almost 1 in 5, experienced at least one form of technology-facilitated sexual exploitation and abuse in a year across 21 countries. Nearly 60 percent of cases occurred on social media platforms, and fewer than 1 percent were reported to authorities.

    Research and evidence; AI-generated abuse imagery · UNICEF

  90. CSAM survivor’s class action alleges xAI trained Grok on child-abuse materialAllegation

    · The Guardian

    A complaint filed in late August in a California federal court alleges that xAI’s Grok generated new illegal sexual images of a survivor using real images of her abuse, and that xAI ingested child sexual abuse material. Musk denied being aware that Grok produced any naked underage images. The allegations are unproven.

    Lawsuits and settlements; AI-generated abuse imagery; AI chatbots and companions · Grok, xAI, AI-generated CSAM, class action, Elon Musk

  91. Character.AI reiterates teen-safety measures

    · blog.character.ai

    On September 3, the company reaffirmed its under-18 open-chat ban, in-house age-estimation model and Internet Watch Foundation membership.

    Legislation and rulemaking; AI chatbots and companions · Character.AI, Internet Watch Foundation, age assurance

  92. Match extends mandatory biometric Face Check to existing Tinder users in US and UK

    · dig.watch

    Match Group began requiring Tinder’s biometric Face Check, a video selfie and facial scan, for existing users in several core markets including the US and UK, according to WIRED. The company says it addresses fake accounts, scams and automated profiles. Security and privacy researchers questioned what the scans prove and warned of normalizing biometric surveillance.

    Privacy and children’s data; Age assurance and verification; Research and evidence · Tinder / Match Group, biometric data, United Kingdom, United States, privacy and data protection

  93. EU Commission escalates rhetoric on Meta and TikTok ‘addictive’ design; 52 MEPs demand action

    · eunews.it

    A letter from 52 MEPs, sent September 2, urged the Commission to act quickly on its DSA investigation of Meta and TikTok after Meta’s US settlement. On September 3, a Commission spokesperson called the companies’ services harmful and addictive for children and said a social media ban proposal is in preparation.

    Lawsuits and settlements; Platform safety features and design; Legislation and rulemaking · Meta, European Commission, TikTok, European Union, European Parliament

  94. Meta details rollout; age assurance to take up to a year, $5.3B contingent on rivals

    · CNBC

    On September 3, Meta said most default teen protections should arrive within about six months and stricter age assurance within up to a year. Meta’s $5.3 billion share of its up to $17 billion settlement is conditional on TikTok and YouTube paying the same amount; Meta also wants them to set one-hour default limits.

    Age assurance and verification · Meta, age assurance, TikTok, YouTube, settlement

  95. Two Tennessee high school students charged over AI-made explicit imagesAllegation

    · WSMV

    The Putnam County Sheriff’s Office said two juveniles used an AI program to create sexually explicit images from photos found online and distributed them. Both face charges of aggravated and especially aggravated sexual exploitation of a minor. The allegations are unproven.

    AI-generated abuse imagery; Education and schools; Criminal cases and prosecutions · criminal prosecution, deepfake nudes in schools, Tennessee (US)

  96. Platforms race to identify child users amid legal pressure

    · pbs.org

    On September 1, an AP analysis details how Meta, Google, TikTok and Roblox are ramping AI age estimation and ID checks, flagging accuracy and bias concerns.

    Age assurance and verification · Google, Meta, Roblox, TikTok, age assurance

  97. LA County child-endangerment suit against Roblox survives anti-SLAPP and dismissal motions

    · mynewsla.com

    On August 31, a Los Angeles Superior Court judge denied Roblox’s anti-SLAPP motion and motion to dismiss in the county’s suit. A hearing to finalize the ruling is set for September 10. The county alleges Roblox marketed itself as safe while exposing children to sexual content, exploitation and predators. The allegations are unproven.

    Legislation and rulemaking; Gaming and virtual worlds; Lawsuits and settlements · Roblox, gaming platforms, sexualized AI content

  98. Dr. Erin Saltman becomes President and CEO of the Tech Coalition, succeeding Sean Litton

    · technologycoalition.org

    The Tech Coalition announced on July 21, 2026 that Dr. Erin Saltman will succeed Sean Litton as President and CEO, starting September 1, 2026. Litton, who led the organization for five years, remained in the role until August 28.

    Industry policy and partnerships; Legislation and rulemaking · Sam Altman, Tech Coalition, international cooperation

  99. Match Group to pay $650,000 in a New Jersey settlement over background-screening disclosures

    · njoag.gov

    On September 1, NJ Attorney General Jennifer Davenport announced that Match Group will pay $650,000 and change its disclosures to settle allegations it violated the state’s Consumer Fraud Act and Internet Dating Safety Act. The allegations involved criminal background screening disclosures across Match, Tinder, Hinge and Plenty of Fish. The allegations are unproven.

    Lawsuits and settlements; Legislation and rulemaking · Tinder / Match Group, settlement, New Jersey (US), consumer protection

  100. Platforms turn to AI age estimation as Meta’s settlement sets age-assurance requirements

    · latimes.com

    An AP analysis published September 1 said age assurance is central to Meta’s $18 billion settlement with 48 states and other child safety measures at Google, TikTok and Roblox. Google and Meta estimate ages with AI from behavior, while Meta deletes ID checks within 30 days. Privacy advocates warn of data risks.

    Age assurance and verification; Legislation and rulemaking; Lawsuits and settlements · Meta, age assurance, settlement, biometric age estimation, Google

  101. Google opens Gemini to K-12 students of all ages in Classroom, catching some schools off guard

    · edweek.org

    In August, Google opened its Gemini chatbot to K-12 students of all ages through Google Classroom, having previously limited it to students 18 and older. The tab appears only in districts that had already allowed Gemini, and administrators can turn it off. Some school leaders said they learned of it from news reports, though Google announced it in June.

    Education and schools; AI chatbots and companions; Legislation and rulemaking · Google, AI in education, Gemini, chatbots