More results...
Without proper security controls, AI agents could perform malicious actions, such as data exfiltration and malware installation.
The post Microsoft Highlights Security Risks Introduced by New Agentic AI Feature appeared first on SecurityWeek.
Why today’s AI attack agents boost human attackers but still fall far from becoming real autonomous weapons. Anthropic recently published a report that sparked a lively debate about what AI agents can actually do during a cyberattack. The study shows an AI system, trained specifically for offensive tasks, handling 80–90% of the tactical workload in […]
India’s AI-powered UAVs are revolutionising warfare by integrating advanced surveillance, reconnaissance, and Kamikaze strike capabilities, marking a significant shift toward technological self-reliance in military operations.Surveillance And Reconnaissance FoundationIndia’s Defence Research and Development Organisation (DRDO) and private firms have developed long-endurance UAVs equipped with
Japan and India have reaffirmed their commitment to deepen cooperation in critical and emerging technologies, particularly semiconductors and artificial intelligence (AI), during a bilateral meeting between Japanese Prime Minister Sanae Takaichi and Prime Minister Narendra Modi on the sidelines of the G20 Summit in South Africa. In this face-to-face meeting,Takaichi expressed her determination to
Prime Minister Narendra Modi met with Netherlands Prime Minister Dick Schoof on the sidelines of the G20 Summit held in Johannesburg on November 23, 2025.During their meeting, the leaders discussed the rapidly growing bilateral partnership between India and the Netherlands, focusing on key sectors such as water resources, innovation, technology, and energy. PM Modi emphasised the commitment of
India’s influence was strongly evident in the G20 South African Summit Leaders’ Declaration, reflecting key priorities from its 2023 G20 Presidency and amplifying Global South concerns.The declaration prominently condemned terrorism "in all its forms and manifestations," marking a firm global stance aligned with India’s push since the Pahalgam terror attack. This condemnation is regarded as one
From Anthropic:
In mid-September 2025, we detected suspicious activity that later investigation determined to be a highly sophisticated espionage campaign. The attackers used AI’s “agentic” capabilities to an unprecedented degree—using AI not just as an advisor, but to execute the cyberattacks themselves.
The threat actor—whom we assess with high confidence was a Chinese state-sponsored group—manipulated our Claude Code tool into attempting infiltration into roughly thirty global targets and succeeded in a small number of cases. The operation targeted large tech companies, financial institutions, chemical manufacturing companies, and government agencies. We believe this is the first documented case of a large-scale cyberattack executed without substantial human intervention…
SquareX claims to have found a way to abuse a hidden Comet API to execute local commands, but Perplexity says the research is fake.
The post SquareX and Perplexity Quarrel Over Alleged Comet Browser Vulnerability appeared first on SecurityWeek.
The company has operated in stealth mode for four months and has signed dozens of customers, including eight unicorns.
The post Runlayer Emerges From Stealth Mode With $11 Million in Funding appeared first on SecurityWeek.
There is a serious security problem inside Comet, the AI-powered agentic browser made by Perplexity, SquareX researchers say: Comet’s MCP API allows the browser’s built-in (but hidden from the user) extensions to issue commands directly to a user…