433 Central Ave., 4th Floor, St. Petersburg, FL 33701 | info@poseidon-us.com | Office: (813) 563-2652

Claude Managed Agents bring execution and control to AI agent workflows

Anthropic’s Claude Managed Agents are a suite of composable APIs for building and deploying cloud-hosted agents at scale, handling sandboxed code execution, checkpointing, credential management, scoped permissions, and end-to-end tracing for you. Developers can define tasks, tools, and permissions within a managed environment, while the platform handles execution and state management. The product is currently available in public beta on the Claude Platform. “Until now, building agents meant spending development cycles on secure infrastructure, state … More → The post Claude Managed Agents bring execution and control to AI agent workflows appeared first on Help Net Security.
http://news.poseidon-us.com/TRy48v

113,000 explicit prompts from AI girlfriend platform exposed, many linked to user IDs

MyLovely.AI, an AI girlfriend platform, suffered a data breach that exposed over 100,000 users. MyLovely.AI allows people to create personalized not safe for work (NSFW) content and engage in real-time conversations with AI-generated companions, often involving highly personal prompts and interactions. According to Have I Been Pwned, the breach exposed email addresses, user-created prompts, links to the resulting AI-generated images, and a limited number of social media profiles, including Discord and X usernames. Based on … More → The post 113,000 explicit prompts from AI girlfriend platform exposed, many linked to user IDs appeared first on Help Net Security.
http://news.poseidon-us.com/TRy47m

Mallory brings contextual threat intelligence to security operations

Mallory is launching an AI-native threat intelligence platform, purpose-built to answer the questions CISOs and their teams are asking every day: What are the real threat vectors for our organization? What’s actually exploitable in our environment right now? What should we proactively fix? The platform monitors thousands of threat sources, contextualizes them against your actual attack surface, and puts that intelligence to work across hunt, detection, and exposure management use cases. One platform. Answers, not … More → The post Mallory brings contextual threat intelligence to security operations appeared first on Help Net Security.
http://news.poseidon-us.com/TRy07n

OPSWAT adds predictive AI engine to MetaDefender for pre-execution threat detection

OPSWAT has announced OPSWAT Predictive Alin AI, its first proprietary AI-based threat detection engine for the MetaDefender Platform. This AI-based innovation introduces a new category of capability within the MetaDefender Platform, a high-confidence predictive layer that works alongside existing detection and prevention engines to assess malicious intent before execution, driving greater efficiency across the platform. This enables organizations to act immediately, while minimizing the operational impacts of false positives. “At OPSWAT, we’ve always believed that … More → The post OPSWAT adds predictive AI engine to MetaDefender for pre-execution threat detection appeared first on Help Net Security.
http://news.poseidon-us.com/TRy07X

Acrobat Reader zero-day exploited in the wild for many months

Unknown attackers have exploited a zero-day Adobe Acrobat Reader vulnerability since November 2025 and possibly even earlier, security researcher Haifei Li has discovered. PDF files carry the exploit Haifei Li is one of the creators of EXPMON, a sandbox-based cybersecurity system for detecting advanced file-based exploits. It does so by analyzing suspicious files submitted through its public web interface or API. “Just few weeks ago, on March 26, someone submitted a PDF sample to EXPMON. … More → The post Acrobat Reader zero-day exploited in the wild for many months appeared first on Help Net Security.
http://news.poseidon-us.com/TRy07T

Intruder expands cloud security with agentless container image scanning

Intruder has announced the release of Container Image Scanning, a new upgrade to its cloud security capabilities that automatically scans container images for vulnerabilities, granting customers actionable insight into container risk without deploying and maintaining scanning agents across their estates. Leveraging existing integrations with major cloud providers, Intruder supports Amazon Web Services Elastic Container Registry, Google Cloud Artifact Registry and Azure Container Registry. New images and updated versions are scanned daily for vulnerabilities, and users … More → The post Intruder expands cloud security with agentless container image scanning appeared first on Help Net Security.
http://news.poseidon-us.com/TRy06w

Advenica’s File Scanner Kiosk scans USB media for malware

Advenica announced the File Scanner Kiosk, a system that scans USB media for malware and helps businesses reduce infection risk. With the reliance on external media for file transfers, organisations face increased vulnerability to malware. The File Scanner Kiosk addresses this challenge by providing an automated, reliable, and efficient way to scan USB media for malware before they enter a secure network. The system uses advanced antivirus technologies to detect and alert users to potential … More → The post Advenica’s File Scanner Kiosk scans USB media for malware appeared first on Help Net Security.
http://news.poseidon-us.com/TRxqvx

Meta’s Muse Spark takes AI a step closer to personal superintelligence

Meta Superintelligence Labs has introduced Muse Spark, a natively multimodal reasoning model with support for tool use, visual chain of thought, and multi-agent orchestration. The release includes a Contemplating mode, which is rolling out gradually and orchestrates multiple agents that reason in parallel. Prompt: Can you turn this into a sudoku game that I can play in the web? (Source: Meta) Capabilities Meta positions Muse Spark as part of its push toward personal superintelligence that … More → The post Meta’s Muse Spark takes AI a step closer to personal superintelligence appeared first on Help Net Security.
http://news.poseidon-us.com/TRxllX

AI agent intent is a starting point, not a security strategy

In this Help Net Security video, Itamar Apelblat, CEO of Token Security, walks through findings from the company’s research, which shows that 65% of agentic chatbots have never been used yet still hold live access credentials. He explains why organizations treat AI agents more like quick experiments than governed identities, and why that creates risks similar to orphaned service accounts, only harder to see. The conversation covers why 51% of external agent actions still rely … More → The post AI agent intent is a starting point, not a security strategy appeared first on Help Net Security.
http://news.poseidon-us.com/TRxbRJ