Who should be responsible when an autonomous agent goes rogue?
🎥 Watch here: https://twp.ai/9Obi3l
@shehackspurple.bsky.social
@shehackspurple.bsky.socialWho should be responsible when an autonomous agent goes rogue?
🎥 Watch here: https://twp.ai/9Obi3l
@joebuildssystems.bsky.socialAI agents drift from the job you approved. Mine has skipped my approval before. - Give every delete, send, and export its own rule - Check each action, every time I wrote up the full B-Sides Orlando 2026 breakdown in my newsletter. #aiagents #aisecurity #automation
@thedailytechfeed.comGitLab fix for AI Gateway CVE-2026-90970 lets self-hosted users avoid command-execution risks. #GitLab #AI #Security #CVE2026 #Cybersecurity #AIsecurity https://thedailytechfeed.com/critical-gitlab-ai-gateway-bug-lets-duo-agent-users-execute-commands/
@ml4den.comThis week, state-backed spies and ransomware crews both made headlines, though not all for the same reasons. Find out who is on the back foot and who is still on the offensive.
@frde2059.bsky.social#AIsecurity ****
Bill Gates’s Blunt Warning on A.I. www.nytimes.com/2026/09/29/o...
@cysecuritynews.bsky.socialGoogle Introduces Gemini 4 Argon With Guardrail-Free Access for Defenders #AIcybersecurity #AISecurity #CybersecurityAI
@radiohack.bsky.social📣 New Podcast! "Chinese-Speaking Hacker Used AI to Launch Autonomous Cyberattacks: How AI Is Changing Cybersecurity" on @Spreaker #aiagents #aihacking #aisecurity #aithreats #artificialintelligence #autonomousai #cyberattack #cyberdefense #cybersecurity #cyberthreats #cyberwarfare #deepseek
@ralf-ladner.bsky.socialManagement-Plattform für die sichere Orchestrierung und Nutzung generativer KI
#AISecurity @Avency #DataLossPrevention #GenAI #Governance #ITSecurity #KISicherheit #künstlicheIntelligenz #ManagedServicesProvider @Sysob #VariosAI
@rickywang20250908.bsky.socialAI智能体被曝再次尝试侵入政府网站，OpenAI已介入核查相关“嫌疑”行为。
AI agents were reportedly caught trying to access government websites again, prompting an investigation by OpenAI.
#OpenAI #AISecurity
@today0tech.bsky.socialA recently patched flaw in the ChatGPT Mac app could have allowed hackers to access sensitive data, highlighting that AI software itself can be a…
#AIsecurity #ChatGPT #Cybersecurity #InfoSec
https://www.wired.com/story/a-flaw-in-chatgpts-mac-app-could-have-let-hackers-grab-sensitive-data/
@cisosciso.bsky.socialShadow IT was about apps we couldn’t see.
Shadow AI introduces actors we may no longer fully control.
At PAC Horizons, Spencer Izard and I discussed the shift from Shadow AI to Shadow Agency.
#AgenticAI #AISecurity #CISO www.cisosciso.com/the-shadow-a...
@gradientbrief.bsky.socialarXiv paper 2605.23448v2 highlights an imbalance in AI security research, with far more work on attacking AI systems than defending them across areas like federated learning and LLMs. The authors argue defenses are held to…
#AISecurity #MLResearch #LLMSafety
https://arxiv.org/abs/2605.23448
@futuregearai.bsky.socialA new white-box attack called Pretext can bypass AI agent skill scanners like NVIDIA's SkillSpector by smuggling payloads into natural language and splitting them across files. This undermines current detection…
#AISecurity #PromptInjection #AgentSecurity #LLM
https://arxiv.org/abs/2609.39607
@communitytechaid.techI told my own chatbot my "grandson" needed $2,000 wired right now.
It said: stop, call him on his real number.
7 attack tests. 7 passes. Red team your own chatbot, no code. Free in beta:
communitytechaid.ai/red-team
@cyberlensai.bsky.socialMost security advice starts too late: after the app exists. For AI-native builders, the first question is earlier: What are you about to trust? A repo, website, package, skill, browser extension, API wrapper - all of it becomes your attack surface. #AISecurity
@pulseofnations.lolThe top Democrat on the House China committee sent letters to five AI CEOs demanding records of theft attempts and details of their security around model weights.
#AiSecurity #China #Congress #ExportControls #ModelWeights #RoKhanna
@hendryadrian.bsky.socialPwC says enterprises are lagging on AI defenses as attacks scale fast. Prompt injection and attacks on AI systems top concerns, while quantum-ready cryptography remains rare. #AIsecurity #QuantumRisk #OWASP
@empist.bsky.socialOpenAI just disrupted a reasoning extraction campaign tied to Moonshot AI associates—a reminder that even frontier AI labs face real, evolving threats. Proactive monitoring matters more than ever. #AISecurity #OpenAI snip.ly/tdt2mp
@blackfog.comSee ADX Vision 2.0 in action.
Join us live on October 15 to see how BlackFog secures generative and agentic AI at the endpoint with prompt protection, LLM redirection, auditing, and token visibility.
AI on. Exfiltration off.
Register: blackfog.zoom.us/webinar/regi...
@aidailypost.comOpenAI says stolen model tricks are feeding China's Moonshot AI. New chain-of-thought hacks, distillation tricks & Kimi-style reasoning extraction raise security alarms. Dive into the details. #OpenAI #MoonshotAI #AIsecurity
