#1 Trusted Cybersecurity News Platform
Followed by 5.20+ million
The Hacker News Logo
Subscribe – Get Latest News
AWS EKS Security Best Practices

Content Moderation | Breaking Cybersecurity News | The Hacker News

Category — Content Moderation
New TokenBreak Attack Bypasses AI Moderation with Single-Character Text Changes

New TokenBreak Attack Bypasses AI Moderation with Single-Character Text Changes

Jun 12, 2025 AI Jailbreaking / Prompt Injection
Cybersecurity researchers have discovered a novel attack technique called TokenBreak that can be used to bypass a large language model's (LLM) safety and content moderation guardrails with just a single character change. "The TokenBreak attack targets a text classification model's tokenization strategy to induce false negatives, leaving end targets vulnerable to attacks that the implemented protection model was put in place to prevent," Kieran Evans, Kasimir Schulz, and Kenneth Yeung said in a report shared with The Hacker News. Tokenization is a fundamental step that LLMs use to break down raw text into their atomic units – i.e., tokens – which are common sequences of characters found in a set of text. To that end, the text input is converted into their numerical representation and fed to the model.  LLMs work by understanding the statistical relationships between these tokens, and produce the next token in a sequence of tokens. The output tokens are detokeni...
U.K. ICO Investigates TikTok, Reddit, and Imgur Over Children's Data Protection Practices

U.K. ICO Investigates TikTok, Reddit, and Imgur Over Children's Data Protection Practices

Mar 03, 2025 Data Privacy / Compliance
The U.K.'s Information Commissioner's Office (ICO) has opened an investigation into online platforms TikTok, Reddit, and Imgur to assess the steps they are taking to protect children between the ages of 13 and 17 in the country. To that end, the watchdog said it's probing how the ByteDance-owned video-sharing service uses the personal data of children in the age range to surface recommendations and deliver suggested content to their feeds. "This is in light of growing concerns about social media and video sharing platforms using data generated by children's online activity in their recommender systems, which could lead to young people being served inappropriate or harmful content," the ICO said . Separately, the data protection regulator said it's also looking into Imgur and Reddit to see how they are using children's information and the measures they are taking to assess the age of their users and tailor content based on that criteria. The ICO ...
Expert Insights Articles Videos
Cybersecurity Resources