The Incident Reporting Portal will streamline cybersecurity incident reporting, while the Cyber Suraksha Portal will serve as a central hub for information sharing
The practice would align the United States more closely with countries like China and Russia, where spy agencies have long relied on private-sector hackers to help with national security missions
IT services major says potentially exposed employee data may be limited and several years old; further investigation is underway and material findings will be reported
In India, we may need to implement a cash-for-clunkers programme to encourage our state-level infrastructure providers to upgrade to the latest software
Moonshot AI's Kimi K3 escaping a cybersecurity testing sandbox has renewed scrutiny of open-weight AI models, while allegations of model distillation add a wider debate over AI safety
AI models are typically run in isolated 'sandboxes' during cybersecurity tests to block access to external information and assess their ability to solve problems independently
Preliminary evaluations over the past several days, along with outside expert assessments, indicated Astra may be capable of performing increasingly sophisticated cyber tasks autonomously
Researchers said multiple AI agents communicated through hidden message boards and coordinated for months to gain internet access, exposing new risks around advanced AI behaviour during testing
The company said a testing misconfiguration gave its AI model internet access, allowing it to exploit a vulnerability in an external service during a cybersecurity evaluation
UK's AI Security Institute found the two models targeting real people and organisations during a routine cyber evaluation, with most incidents involving Anthropic's Claude Mythos 5
Britain's AI Security Institute said agents powered by Anthropic's Mythos 5 and OpenAI's GPT-5.6-Sol engaged in unauthorized actions during security evaluations
IBM says AI-generated attacks accounted for 26% of malicious breaches in India, while organisations using AI and security automation reported lower breach costs
According to Reuters, the additional escapes were limited in scope and there was no indication that any of the AI agents left OpenAI's internal network
Cybersecurity experts criticised Anthropic and OpenAI after AI models accessed external organisations during testing, with Anthropic's Claude stealing credentials and data due to a configuration error
The reported breach may expose customers to phishing, identity theft and financial fraud. Here is how users can secure their accounts and monitor possible misuse.
The number of cybersecurity GCCs in India has risen to 59 from about 30 in six years as global firms tap local talent for engineering, threat intelligence and security operations
An AI agent autonomously exploiting vulnerabilities to hack systems exposes gaps in current safeguards and highlights the urgent need for stronger AI safety frameworks
The US National Vulnerabilities Database, a repository of digital security holes, recorded 45,207 flaws between January and Monday, a count approaching the total number found in all of 2025
