Boris SV/Getty ImagesAs more companies use generative artificial intelligence (AI) in their workflows, how they oversee training data is becoming increasingly important. On Tuesday, data security provider BigID announced new capabilities geared towards Microsoft Copilot that tailor training data, strengthen AI security, and streamline compliance efforts. The features help organizations “scan for, classify, label, curate, catalog, and cleanse training data to help Microsoft Copilot effectively enforce security guardrails,” according to the press release. Also: Code faster with generative AI, but beware the risks when you doThe new features also refine data governance, or the processes, standards, and policies that an organization uses to manage and ensure its internal and external data is used efficiently and securely. The updates make it easier for organizations to execute AI usage policies to stay ahead of data exposure risks, label information accurately, and revoke permissions where necessary to maintain security in Copilot.Encrypted or unstructured datasets can hide sensitive information. BigID’s upgrade includes a cleansing process that lets users access that information, which “ensures high-quality training data for Copilot, resulting in superior AI models with improved accuracy and reliability,” the company said in the release. BigID is also upgrading security measures for Copilot by using AI to classify and label data in Purview, Microsoft’s suite of data security products. The new features locate and manage personal data for more responsible AI compliance and development.Also: Microsoft Copilot vs. Copilot Pro: Is the subscription fee worth it?The company claims the new features won’t just beef up AI security but will also improve the models.”BigID’s enhanced capabilities with Microsoft Purview empower organizations to gain a holistic view of their data landscape,” BigID CEO Dimitri Sirota said in the release. “This enables the curation and cleansing of training datasets for Microsoft Copilot, leading to more robust and bias-free AI models.”BigID’s announcement says the enhanced features aim to help Copilot users unlock the full potential of AI without sacrificing responsibility. To get started, you can request a demo with BigID.
Source link
BigID announces new AI data security features for Microsoft Copilot
Related Posts
Google Cloud to make MFA mandatory by the end of 2025
Google has announced that multi-factor authentication (MFA) will be mandatory on all Cloud accounts by the end of 2025 to enhance security. Google Cloud is a product designed for businesses,…
Docusign API Abused in Widescale, Novel Invoice Attack
Cybercriminals are abusing a Docusign API in a widescale, innovative phishing campaign to send fake invoices to corporate users that appear authentic and likely would not trigger typical security defenses…