Microsoft (MSFT, Financials) is expanding its cybersecurity offerings with new artificial intelligence-powered tools to address a sharp rise in digital threats and the challenges of securing generative AI.
The company said on Monday that it will be releasing 11 new Security Copilot agents in April. These agents will help automate a lot of security jobs, such as finding phishing attempts, stopping data loss, managing vulnerabilities, and controlling who can access your information.Microsoft says there have been a lot more hacks lately. Every day, its systems look at 84 trillion security signs and stop over 30 billion fake emails until 2024. The company says that these levels of traffic are too much for security teams that use human methods for finding and sorting messages.It took six months of work to make six of the agents. They include tools for phishing review, insider risk management, identity access policy enforcement, vulnerability repair, threat intelligence gathering, and more. The tools are made to work with all of Microsoft’s systems, including Microsoft Defender, Microsoft Purview, and Microsoft Entra. This will help security teams respond faster, respond to risks in real time, and improve operations while keeping control.Microsoft is also delivering five more agents with the help of partners such as OneTrust, Aviatrix, BlueVoyant, Tanium, and Fletch. These solutions will help businesses handle data breaches, keep an eye on network problems, evaluate security operations, put danger alerts in context, and decide which big cyber risks need the most attention.Microsoft said it will roll out controls to find and stop illegal access to shadow AI apps, which are unapproved creative AI tools that are used without IT oversight. This comes as worries about these apps grow. Some of these are a new AI web category screen and rules built into Microsoft Edge for businesses to stop data loss. These features are meant to keep private data from being entered into apps like Google Gemini, ChatGPT, Copilot Chat, and DeepSeek.The company is also adding more cloud providers and AI models to its AI security posture management. Microsoft Defender will protect Google Vertex AI, Meta Llama, Mistral, and other models in the Azure AI Foundry starting in May 2025. The Open Worldwide Application Security Project lists threats like indirect prompt injection attacks, private data leaks, and token abuse. More tools will be made available to find these threats.Microsoft said that starting in April 2025, one of its new features will be hacking defense in Microsoft Teams. Defender for Office 365 will also offer real-time link destruction and warning viewing.The improvements are part of the company’s larger effort to help secure AI change as more businesses use creative tools. Microsoft pointed to its own study that showed 57% of companies have had more security problems since using AI, while 60% have not yet put controls in place.
This article first appeared on GuruFocus.
发表回复