Security
-
Dutch software links air defence drones for Ukraine
Dutch company Intelic has demonstrated software designed to connect radar, detection systems and intercepting drones for air defence. Its Nexus platform is being used in Ukraine, while the Netherlands has a €30m defence contract with the company for related capabilities.
-
RSA gives AI agents identities of their own
RSA is extending identity controls to AI agents, creating records of ownership and permissions while applying policy to individual tool calls as regulated organisations move autonomous software into sensitive systems.
-
Instagram Instants tests Ofcom’s product change rules
Ofcom is investigating whether Meta carried out adequate illegal content and child safety risk assessments before launching Instagram Instants, testing Online Safety Act duties that apply when platforms make significant changes to their services.
-
Netsec collapses the boundary between IT and security
Netsec has raised more than $10 million for a managed service that combines IT operations and cybersecurity under one provider, using automation to handle routine administration while human operators retain responsibility for incidents and exceptions.
-
Hadrian puts more capital behind continuous offensive testing
Hadrian has raised $40 million to expand a security platform that continuously discovers internet facing assets and uses AI agents to test which weaknesses can actually be exploited before attackers find them.
-
AI bug reports overwhelm Google’s open source bounty
Google has stopped accepting new product vulnerability submissions through its open source bug bounty programme after automated reports rose sharply and the vast majority proved invalid, exposing a new bottleneck in AI assisted security research.
-
Fleuret puts AI agents on the offensive side of cyber security
Paris startup Fleuret has raised €4 million to automate penetration testing with AI agents that discover systems, identify weaknesses, attempt exploitation and retest fixes as infrastructure changes.
-
Denmark breach exposes risk in trusted access
Attackers abused a private company’s legitimate access to Denmark’s population register to obtain personal data linked to about 8.8 million registered people, exposing the risk created when third parties connect to critical public data.
-
AI agents widen the enterprise attack surface
Microsoft’s latest defence report argues that AI agents introduce security exposure through the identities, data, tools, and systems they are authorised to use.
-
AI access becomes a cybercrime commodity
CERT-EU and Google threat research show attackers stealing AI accounts and hijacking cloud infrastructure to obtain expensive models and computing capacity.
