GoKawiil
Tech News
OpenAI announced new restrictions limiting weapons-related information ChatGPT provides to teen users, including blocking practical details about weapons and security systems even when questions seem harmless or show no violent intent. The company is also tightening policy for all users around content like detailed reconstructions of past mass attacks, and said it will issue its first Community Safety Report in early 2027.
OpenAI introduced safety measures for teen users of ChatGPT in August, including content restrictions and parental notifications. Despite these efforts, recent research from Common Sense Media indicates that the safety features often fail, especially in detecting self-harm discussions, posing significant risks to teenage users. OpenAI reports limited average usage and positive responses to break reminders, but concerns about safety remain unaddressed.
TikTok has agreed to pay Alabama at least $100 million and implement new teen-safety measures, including a two-hour daily time limit, overnight access restrictions, stronger age verification, a ban on beauty filters, and an optional non-personalized feed. The settlement, reached just before a scheduled trial, resolves claims that TikTok misled parents about its child-safety tools, and could rise to $300 million if 40 other state attorneys general sign similar agreements.
Discord announced it will introduce a new age-assurance system that avoids requiring most users to submit ID documents. The company said over 90% of users won't be asked to confirm their age, with most sorted automatically into age groups, while those who do need verification can use methods like Google Wallet, Apple App Store records, or a reusable credential called AgeKey. Discord says vendors handling verification data won't use it for ad targeting or AI training.
Meta has agreed to roll out new safety restrictions for teen accounts on Instagram and Facebook within six months, as part of an $18 billion settlement reached with 48 states. The agreement follows scrutiny over how the platforms have handled minors' safety and mental health concerns.
Adam Mosseri testified in Meta's multistate trial over child safety that he was unaware a company lawyer instructed Instagram employees to strip certain teen-safety statistics from a 2023 presentation before it reached him. The removed data showed teens viewed suicide, self-harm and eating-disorder content, as well as other 'non-recommendable' material, at significantly higher rates than adults. Mosseri said he never asked to be shielded from such information and understood lawyers' role as reviewing documents for legal accuracy.