Влияние ИИ на кибербезопасность: MITRE ATLAS и новый ландшафт угроз

Сегодня искусственный интеллект кардинально меняет как подходы к защите, так и методы атак. С развитием технологий ИИ-модели могут обрабатывать и анализировать огромные объемы данных в реальном времени. Это активно использует не только бизнес, но и злоумышленники. В статье рассмотрим современные методы атак на AI и ML-системы, расскажем про практическое применение MITRE ATLAS для моделирования угроз и выстраивания защиты через четыре системных элемента: AI Среда, AI Платформа, AI Модель и AI Данные.

https://habr.com/ru/companies/infera_security/articles/1046568/

#AI_Firewall #MITRE_ATLAS #prompt_injection #SAFEAI #защита_AI_моделей #безопасность_ИИ

This week was all about Anthropic:

Their CEO Dario Amodei was in Australia to sign an MOU with the Australian Government.

They also unintentionally exposed some of Claude Code’s source code, which they tried to pull down using copyright takedown notices (just another reminder of the hypocrisy of AI).

https://elliottbledsoe.wtf/wtf-now-29/

#Anthropic #Claude #SafeAI #Takedowns

🤖 Safe Learning Systems & AI
How can AI stay reliable under uncertainty? At the AI Colloquium, Prof. Nils Jansen explains how formal methods and reinforcement learning contribute to trustworthy AI. 📐🧠

📅 15 Jan 2026 | ⏰ 10:15 AM
📍 TU Dortmund University, JvF 25, 3rd Floor

💬 What makes AI trustworthy in your view?

Photo: Ruhr University Bochum

#AI #SafeAI #TrustworthyAI #MachineLearning #FormalMethods #Research #AIColloquium

Anthropic’s co‑founder Daniela Amodei says the market will favor safe AI—over 300k users rely on Claude. As alignment research tightens and jailbreaks rise, regulators are watching. Can transparent deployment keep the edge? Read how safety could become a competitive advantage. #Anthropic #ClaudeAI #SafeAI #AIAlignment

🔗 https://aidailypost.com/news/anthropics-amodei-says-market-will-reward-safe-ai-300000-use-claude

🤖 Smart AI needs smarter boundaries.

AI is reshaping our world—but without ethics, innovation can become a threat.
⚖️ Let’s define the line between progress and responsibility. Read more!👇

https://neuronus.net/en/blog/ethical-AI-threats-and-solutions

#AIethics #EthicalTechnology #AILegislation #AIandSociety #AIrisks #DigitalResponsibility #AIgovernance #SafeAI #Neuronus

AI is rising fast—but can we keep it in check?

As AI reshapes our world, global regulation is key to ensuring it serves humanity—not harms it.
🧭 Let’s guide AI with smart, united regulation.

🔍 Learn more about AI laws—read the full blog.👇

https://neuronus.net/en/blog/ai-legislation-and-regulations

#AIRegulation #AIandLaw #ResponsibleAI #EthicalAI #SafeAI #TechRegulation #GlobalTechPolicy #Neuronus

Yoshua Bengio Launches LawZero: A New Nonprofit Advancing Safe-by-Design AI

https://lawzero.org/en/news/yoshua-bengio-launches-lawzero-new-nonprofit-advancing-safe-design-ai

#HackerNews #YoshuaBengio #LawZero #SafeAI #Nonprofit #AIethics #Innovation