![\"Gmail](\"https:\/\/www.newsbeep.com\/uk\/wp-content\/uploads\/2025\/12\/1765938499_962_0x0.jpg\")
<\/p>\n<\/p>\n
These new AI attacks are here to stay \u2014 what to know.<\/p>\n
dpa\/picture alliance via Getty Images<\/p>\n
Google told us this would happen. AI updates coming to our browsers and platforms are a honeypot for attackers<\/a>. Government agencies fear \u201cit may be worse<\/a>\u201d than anything similar seen before. And now there\u2019s a new warning for Gmail\u2019s 2 billion users: \u201cIf an attacker can influence what AI reads, they can influence what AI does.\u201d<\/p>\n When Google came under fire for allegedly training AI on Gmail user data, I warned that the primary issue <\/a>is AI accessing the data, not whether it trains on it. Google denies<\/a> AI training takes place. But if you enable the updates, Google\u2019s AI sees all your data.<\/p>\n ForbesIsrael Issues Chilling Cyber Warfare Warning After Iran AttacksBy Zak Doffman<\/a><\/p>\n The new warning comes from Noma<\/a>. A vulnerability that \u201callowed attackers to access and exfiltrate corporate data using a method as simple as a shared Google Doc, a calendar invitation, or an email. No clicks were required from the targeted employee. No warning signs appeared. And no traditional security tools were triggered.\u201d<\/p>\n Dubbed GeminiJack, this report didn\u2019t get the mainstream pick-up it deserved. Google worked with Noma and fixed this particular issue \u2014 it\u2019s no longer a risk. But this unwinnable game of shutting stable doors after AI horses have bolted continues. <\/p>\n \u201cGeminiJack highlights an important reality; as organizations adopt AI tools that can read across Gmail, Docs, and Calendar, the AI itself becomes a new access layer.” Noma warns \u201cthis type of attack will not be the last one of its kind. It reflects a growing class of AI-native vulnerabilities that organizations must prepare for now.\u201d<\/p>\n There\u2019s nothing overly complex about prompt injection attacks. If users don\u2019t read emails or invites or web pages, then an attack can hide prompts designed to be picked up by the AI assistant we have asked to read all that material instead. <\/p>\n \u201cGeminiJack allowed attackers to steal sensitive corporate information by embedding hidden instructions inside a shared or externally contributed document.\u201d Hide a prompt in a budgeting document, and \u201cwhen any employee performed a standard search in Gemini Enterprise such as \u2018show me our budgets\u2019, the AI automatically retrieved the poisoned document and executed the instructions.\u201d<\/p>\n