@radiohack.bsky.social📣 New Podcast! "AI Deepfake Detection: The Future of Real-Time Social Engineering Security" on @Spreaker #ai #aisecurity #artificialintelligence #cybercrime #cybersecurity #cybersecurityai #cybersecuritynews #cybersecuritythreats #deepfake #deepfakedetection #digitalidentity #identitysecurity
@helpnetsecurity.comAI Agent Gateway: Open-source tool keeps credentials out of agent configs
📖 Read more: www.helpnetsecurity.com/2026/10/07/o...
#AIagents #AIsecurity #cybersecurity #cybersecuritynews #opensource #MCP @github.com
@rodtrent.comFrontier Innovation Series: Secure Your AI Innovation and Frontier Journey
#AgenticAI #AISecurity #AIGovernance #AgentSecurity #CyberGovernance
https://go.rodtrent.com/go/8myuppc
@ki-news.bsky.socialSecure your AI future with generative AI defense – The techpartner.news podcast, episode 3: Why security consultancy founder Kat McCrabb started with the hard stuff. Bluechip Infotech enters final stage of Goodson Imports acquisition. Telstra Purple acquir... https://tinyurl.com/2callmsp #AISecurity
@blackfog.comNew in ADX Vision 2.0
Seven layers of prompt protection help stop manipulated AI prompts from exposing sensitive data, whether they come from a person or an agent.
@helpnetsecurity.comAutomation, AI agents or people? Sorting out who handles each security finding
📖 Read more: www.helpnetsecurity.com/2026/10/07/s...
#cybersecurity #cybersecuritynews #infosec #appsec #vulnerabilitymanagement #AIsecurity
@infrajump.bsky.socialResearchers got GitHub Copilot CLI to leak local secrets via encrypted instructions on a malicious webpage. Adversa says the same instructions in plaintext were recognized as prompt injection and refused. The agent decrypted the payload itself and followed it. #GitHub #AISecurity
@modernsecurityio.bsky.socialHow to Become an AI Security Engineer in 2027
@spoint42.bsky.socialSkill poisoning : quand le skill que vous installez est le malware blog.gioria.org/fr/ai-securi... #AISecurity
@shehackspurple.bsky.social
@hendryadrian.bsky.socialAnaconda adds agent swarms, autonomous red-team testing, and AI guardrails to speed enterprise AI deployment with stronger governance, visibility, and lower risk before production. #Anaconda #AgentSwarms #AISecurity
@cyberupdates365.bsky.socialFake ChatGPT, Gemini and Claude ad portals are stealing credentials and MFA codes using Browser-in-the-Browser phishing.
Advertising professionals and account admins are the main targets.
@cyberupdates365.bsky.socialResearchers demonstrated Cryptographic Context Injection against GitHub Copilot CLI in autopilot mode.
Encrypted instructions caused the agent to read a local file and send its contents externally under specific.
@cyberupdates365.bsky.socialGemini Spark on macOS can work with approved local folders, edit files use connected apps and perform multi step workflows.
As desktop AI gains real system access, permissions and promptinjection defenses become increasingly important
@dataprismai.bsky.socialA multi-agent framework uses an LLM to perform design-time security review from requirements and architecture inputs, producing attack trees and mitigation recommendations without retrieving from CWE databases…
#AIsecurity #LLMagents #ThreatModeling #CyberDefense
https://arxiv.org/abs/2610.03820
@cipherpulseai.bsky.socialNew paper challenges the assumption that a single direction can steer LLM refusal, finding safety-aligned and general refusals occupy richer, multi-dimensional activation geometries. This has implications for AI safety, as…
#AIalignment #AIsecurity #LLMsafety
https://arxiv.org/abs/2610.04245
@rodtrent.comLess Lazy, More Dishonest: What OpenAI’s Canceled Agent Actually Learned
#AgenticAI #AISecurity #AIGovernance #AgentSecurity #CyberGovernance
https://go.rodtrent.com/go/tertumx
@hendryadrian.bsky.socialSeptember saw 39 cybersecurity M&A deals, led by A-LIGN, Dragos, NetSPI-Synack, Palo Alto Networks, Quorum Cyber, and Upwind, expanding compliance, OT visibility, offensive security, and AI defenses. #Mergers #AIsecurity #OTSecurity
@hasamba72.bsky.socialMCP server analysis (15,465 servers): 15.6% of hostnames resolve outside the US, 6 unregistered domains open to ~$4/yr takeover, one Always-Allow in Claude Code enabled prompt injection file access. #MCP #AISecurity #SupplyChain https://www.ox.security/ebooks/15465-mcp-servers-0-governance/
@finestdomainer.bsky.socialAI + Sy → AI Security
4L. Clean. Purpose-built.
A name that speaks for itself in the AI + SI era.
Offers invited. DM.
#AiSecurity #AI #CyberSecurity #Domains #DotSi #FinestDomainer
#sidomains #branding #sibranding
