Campus Technology · সাধারণ

Sophos Launches Tool to Defend Against Shadow AI

২৬/১০/২০২৬

Sophos Launches Tool to Defend Against Shadow AI

Unauthorized artificial intelligence tools and agents are creating security risks for enterprises as AI adoption increases, according to cybersecurity company Sophos. In a recent product announcement, the company highlighted the growing use of shadow AI, referring to AI tools and agents operating without formal organizational approval or oversight. The security firm cited research indicating that 69% of organizations suspect or have evidence of unauthorized AI use.

Separate research from Gartner and Cloud Security Alliance found that 82% of organizations discovered shadow AI agents in their environments during the previous year.

Unauthorized AI Creates Security Visibility Gaps

According to the security firm, organizations have struggled to identify which AI tools employees use, what information those tools access, and whether their use complies with security policies.

"Organizations are adopting AI faster than they can secure it, and most security teams can't answer three basic questions: which AI tools are running, what data is going into them, and whether that activity lines up with policy," said Kyle Falkenhagen, senior vice president of product management at Sophos.

The company also cited an IDC projection that the number of AI agents could reach 1.3 billion by 2028.

Sophos researchers reported observing attackers use AI agents within a customer's network to build and test nearly 80 malware modules.

Sophos Launches AI Defense

The cybersecurity firm announced the general availability of Sophos AI Defense, a security offering designed to identify AI tools, monitor their activity, and enforce organizational security policies.

The product operates through Sophos Fusion, the company's cybersecurity platform, and uses information from devices, networks, identity systems, and browsers.

According to the company, AI Defense can inspect prompts, uploads, and other AI interactions to identify potentially risky activity and help prevent sensitive data exposure.

For more information, visit the Sophos site.

Featured

Cloud Security Alliance Recommends Separate Governance Model for 'Non-Human Identities'

A recent report urges organizations to inventory, govern, and continuously monitor machine and agent identities.
OpenAI's Astra Model Reaches Critical Cyber Threshold

OpenAI has unveiled GPT-6 Astra, its most capable broadly deployed model and the first system the company says has reached the "Critical" cybersecurity capability threshold under its Preparedness Framework.
Anthropic Expands Enterprise Deployment Options for Claude Desktop with New Controls and Cloud Integrations

Anthropic is adding new enterprise deployment options for Claude Desktop, saying organizations that use the app through Amazon Web Services, Google Cloud, and Microsoft Foundry can now access the full desktop experience across chat, Claude Cowork, and Claude Code.
Claude Sonnet 5.5 Offers Model-Switching Cybersecurity Safeguards

Anthropic's latest Claude release, Sonnet 5.5, brings cybersecurity safeguards that can change which model handles a request during a Claude conversation. Higher-risk tasks can fall back to Sonnet 5.