This week saw major steps in AI-driven security tools and privacy laws impacting tech companies. Microsoft launched Security Copilot agents to handle phishing attacks, freeing human teams for complex tasks. The EU introduced stricter rules under the EDPB to control how AI uses scraped data, aiming to protect user privacy. In the U.S., OpenAI partnered with National Labs to boost nuclear security using AI, while Chinese firm DeepSeek caused stock drops with its advanced AI app.

Companies like Cloudera and AWS reported that compliance delays are slowing AI projects, as rules vary globally. A survey found security and data privacy as top software challenges, with 51% of tech leaders struggling to secure AI systems. Meanwhile, Microsoft added Teams protection against phishing and controls to block sensitive data leaks into unapproved AI apps.

Europe’s AI Act pushed firms to audit tools for bias and privacy risks, while the U.S. debated exemptions for national security AI. Experts urged clearer guidelines to balance innovation and safety, as shadow AI (unofficial apps) raised leakage risks.