Microsoft is launching 11 AI cybersecurity “agents” designed to combat the increasing complexity of hacking attempts by criminals, fraudsters, and spy agencies. The move comes as a surge in phishing emails has reached unprecedented levels, with 30 billion emails tracked last year.
According to Vasu Jakkal, vice president of security at Microsoft, humans cannot keep up with this volume, making AI-powered solutions essential. The new agents will identify and sift through suspicious emails, block hacking attempts, and gather intelligence on attack origins.
With Windows running on around 70% of the world’s computers, Microsoft is a prime target for hackers. The company has seen a five-fold increase in organised hacking groups, with a “gig economy” of cybercriminals worth $9.2trn (£7.1trn).
Microsoft’s AI agents will be incorporated into its Copilot portfolio and serve IT and cybersecurity teams rather than individual Windows users. However, some experts are concerned about the potential risks of autonomous AI agents on vulnerable networks.
To mitigate these concerns, Microsoft applies a “zero trust framework” to its AI tools, constantly assessing whether agents are playing by their programmed rules. The company’s release of multiple cybersecurity agents ensures each AI has a defined role and only accesses relevant data.
Source: https://news.sky.com/story/microsoft-looks-to-ai-to-close-window-on-hackers-13335086