Threats

• The EU is investing €1.3 billion (approximately $1.5 billion) into AI, cybersecurity, and digital skills via the Digital Europe Programme from 2025 to 2027. • New regulatory frameworks, specifically the NIS2 Directive and the Digital Operational Resilience Act, became effective in October 2024 and January 2025. • These mandates make cybersecurity investments a legal requirement for critical sectors, including transport, energy, and financial services, to protect essential infrastructure.

Nauran government says its citizens are ‘friendly’ after MP spoke of serious threats allegedly made against former Australian detaineesGet our breaking news email, free app or daily news podcastThe Nauruan government has issued a rare statement insisting it is a “friendly” and “welcoming” country after a whistleblower alleged “serious threats of physical violence” were made against a group of non-citizens removed there by the Albanese government.The unexpected defence, sent shortly after midday Thursday, was mounted hours after independent MP Andrew Wilkie used his three-minute constituency statement to read claims from an anonymous whistleblower familiar with the arrangements of the secretive $2.5bn deal between Australia and Nauru. Continue reading...

• Sam Woods, the outgoing chief of the Prudential Regulation Authority (PRA), has identified AI-driven cybersecurity risks as the top threat facing the banking sector. • The regulator expressed significant concern regarding existing vulnerabilities within lenders' IT systems, which could be exploited by sophisticated AI tools. • This warning highlights a critical shift in the threat landscape, as AI enables more complex and frequent cyberattacks against financial infrastructure.

• AI-powered cyber threats are projected to become significantly more autonomous and sophisticated by 2026. • These advanced attacks are outstripping traditional security frameworks, allowing threat actors to bypass conventional defenses with greater ease. • This shift matters because AI fundamentally reshapes the cybersecurity landscape, increasing the danger and scale of digital incursions.

• Many companies are failing to secure their systems as they rapidly integrate Large Language Models (LLMs) and generative AI. • Security gaps are being driven by "shadow AI" and weak internal controls, leaving firms vulnerable to data breaches, hallucinations, and unauthorized access. • This lack of preparation is critical because cybersecurity threats have increased exponentially in recent years, outpacing corporate defense strategies.

• Policy makers are currently working to establish a coordinated AI-security regime focused on frontier models, critical infrastructure, and voluntary government reviews. • Simultaneously, cyber attackers are evolving their tactics by refining malware distribution and abusing familiar platforms to target ordinary users at scale. • This tension highlights a growing gap between high-level government security frameworks and the rapid proliferation of low-friction criminal tooling.

With corrupt police on the streets and shopkeepers forced to pay gangs, president has vowed to tackle crime that now affects all parts of societyIt was about 11pm and Luis* was about to get into an Uber to go home when the police car pulled up. One of the officers produced two plastic bags with what looked like drugs: one contained some sort of powder, the other little crystals. Luis had never seen them before.Luis, who asked not to use his real name for fear of reprisals, insisted that the drugs weren’t his, but the officers didn’t seem to care. They shoved him into the back of the police truck and drove into the night. Continue reading...

• Microsoft is integrating artificial intelligence into its security operations to better combat evolving cyber threats and hacker activity. • The strategy utilizes an AI model revealed by Anthropic in April, which has demonstrated a high proficiency in identifying vulnerabilities within major operating systems and web browsers. • This development is significant because the AI was not originally designed for cybersecurity, yet it can uncover critical software flaws that traditional methods might miss.