Security News Roundup: LastPass Breached Again, Bolton Pleads Guilty, Infostealers Disrupted

This week saw several significant cybersecurity incidents.

LastPass Breach via Partner

Password manager LastPass notified customers of a data breach resulting from an attack on AI business intelligence firm Klue. Attackers compromised access tokens for Klue customers, including LastPass, and used them to extract customer names, phone numbers, email addresses, physical addresses, support case data, and sales information from Salesforce and other integrated platforms. LastPass emphasized that its own infrastructure was not compromised and password vaults were unaffected. The company urged vigilance against phishing attacks leveraging exposed contact details.

John Bolton Pleads Guilty

Former US National Security Advisor John Bolton pleaded guilty Friday to one count of illegal retention of classified defense information. Bolton, 77, struck a plea deal that may allow him to avoid prison, though it recommends up to five years. Sentencing is set for October 28 before US District Judge Theodore Chuang. The deal also includes a $2.25 million fine, with Bolton allowed to withdraw his plea if the sentence exceeds the recommendation.

International Operation Targets Infostealers

Microsoft, Europol, and other partners announced Wednesday the disruption of Amadey and StealC infostealer malware infrastructure under Operation Endgame. The action seized 326 servers and 142 domains, flagged approximately $47 million in stolen cryptocurrency, and recovered up to 27 million stolen credentials. Microsoft used AI-assisted analysis to show both malware relied on the same backend infrastructure.

Australia Discovers State Hackers in Critical Infrastructure

Australia's Security and Intelligence Organisation (ASIO) reported finding nation-state hackers inside a critical infrastructure provider's network. The hackers had obtained credentials for active users, including IT professionals, and were preparing for sabotage. ASIO has established dedicated teams to counter such attacks.

Other News

In Bristol, England, a predictive policing program has used 23 models over a decade to assess crime risks. Members of Peter Thiel's private group Dialog had their personal data exposed due to a website misconfiguration. Anthropic received permission to release its Mythos 5 model to select US companies and agencies. OpenAI launched an improved GPT-5.5-Cyber model and a "Patch the Planet" initiative to support open-source vulnerability patching. World Cup-related scams are becoming harder to detect.