Can Meta’s AI Scam Detector Actually Stop Them?

This week on The Awareness Angle, Ant Davis and Luke Pettigrew unpack a wave of global cyber stories — from telecom breaches and AI-powered defence tools to sextortion scams and the emotional risks of “friendly” chatbots. It’s a mix of human stories, technical takeaways, and practical lessons for anyone trying to stay safe in an AI-shaped world.📡 Global Breaches & Third-Party Fallout – LG U+, Toys “R” Us Canada, HSBC, and Verisure all suffer breaches linked to vendors or poor visibility. The takeaway? Even mature orgs keep getting blindsided by supplier access and delayed disclosure.🤖 OpenAI’s ‘Aardvark’ GPT-5 Agent – A self-fixing AI for security flaws sounds promising—until you realise it’s patching live code. Automation helps, but trust and verification still matter more than ever.💬 Meta’s Scam Detector – WhatsApp and Messenger now use AI to flag impersonation and job scams. Ant ties this to his own “Tilly from Fram Search” scam attempt, showing how emotional hooks still trump logic.🧒 AI Sextortion Scams & ReportRemove – Deepfaked nudes used to extort teens; a BBC case highlights the IWF’s lifesaving removal tool. A reminder that awareness isn’t just about security—it’s safeguarding.👥 Character.AI Blocks Teen Chat – After reports of inappropriate AI conversations, under-18s are now cut off. Ant and Luke discuss why “empathetic” AI companions can quickly turn toxic.🇬🇧 NCSC Annual Review – Four major UK cyber incidents every week, a 129% rise year-on-year. New SME Cyber Action Toolkit promises easy wins, but small firms still face time and funding barriers.🧩 Chrome Zero-Day (Memento Mori) – Active exploit patched, but only if users reboot. Awareness message: “Auto-update isn’t a shield—restart and verify.”💼 Insider Threats & Classroom Tricks – A Reddit post shows real insider exfiltration, while teachers hide invisible AI prompts to catch students using ChatGPT. Both show behaviour—not tech—is the true battleground.📰 AI Authenticity Crisis – From AI-written beauty magazines to GPT vs Google explainers, even “real” media now demands literacy training to spot synthetic content.🧠 ‘EtherHiding’ Malware on Blockchain – Malicious code hidden in blockchain assets targets job seekers via fake coding tests. Proof that persistence now has a whole new meaning.Whether you’re defending systems, teaching staff, or just trying to keep your kids safe online—this episode connects the technical, the human, and the emotional sides of cybersecurity.🕒 Timestamps00:00 — Introduction & Milestone Celebration📩 For links, videos, and the newsletter – head to ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠riskycreative.com⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠💬 ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Check Out This Episode's Discussion Points⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠📧 ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠[email protected]⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠🔗⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ riskycreative.com⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠🎵 Our Intro & Outro Song (© ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠16! by ⁠falling forever⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠)License: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://creativecommons.org/licenses/by/4.0⁠⁠⁠