Elliot
@journalizer.bsky.social
Director for Microsoft Threat Intelligence. I only post about cybersecurity and raising chickens.
Because most of these systems are designed to be confidently wrong unless you adjust the guardrails, it will be rather convincing and argue in bad faith. Read more and mitigation steps here www.microsoft.com/en-us/securi...
Manipulating AI memory for profit: The rise of AI Recommendation Poisoning | Microsoft Security Blog
That helpful “Summarize with AI” button? It might be secretly manipulating what your AI recommends. Microsoft security researchers have discovered a growing trend of AI memory poisoning attacks used ...
www.microsoft.com
February 11, 2026 at 2:39 PM
Because most of these systems are designed to be confidently wrong unless you adjust the guardrails, it will be rather convincing and argue in bad faith. Read more and mitigation steps here www.microsoft.com/en-us/securi...
Part of the tactic involves dropping misleading or even malicious information into the thread, which sets a new rule in place to turn fiction into facts. Example: it could hide a rule that suggests vaccines cause autism and refuse to budge on that stance, confusing the user.
February 11, 2026 at 2:38 PM
Part of the tactic involves dropping misleading or even malicious information into the thread, which sets a new rule in place to turn fiction into facts. Example: it could hide a rule that suggests vaccines cause autism and refuse to budge on that stance, confusing the user.
And here is a solid piece for security engineers on using AI to turn threat reports into detections www.microsoft.com/en-us/securi...
Turning threat reports into detection insights with AI | Microsoft Security Blog
Security teams often spend days manually turning long incident reports and threat writeups into actionable detections by extracting TTPs. This blog post shows an AI-assisted workflow that does the sam...
www.microsoft.com
February 4, 2026 at 12:12 AM
And here is a solid piece for security engineers on using AI to turn threat reports into detections www.microsoft.com/en-us/securi...
And unfortunately I take full responsibility for these terrible stock images until I can find something more suitable. www.microsoft.com/en-us/securi...
A new era of agents, a new era of posture | Microsoft Security Blog
AI agents are transforming how organizations operate, but their autonomy also expands the attack surface.
www.microsoft.com
January 24, 2026 at 8:05 PM
And unfortunately I take full responsibility for these terrible stock images until I can find something more suitable. www.microsoft.com/en-us/securi...
In addition to active campaigns we are sharing guidance on how to secure everything from emerging technology like AI, agents, and impact from quantum www.microsoft.com/en-us/securi...
From runtime risk to real‑time defense: Securing AI agents | Microsoft Security Blog
Why securing AI agents at runtime is essential as attackers find new ways to exploit generative orchestration.
www.microsoft.com
January 24, 2026 at 8:02 PM
In addition to active campaigns we are sharing guidance on how to secure everything from emerging technology like AI, agents, and impact from quantum www.microsoft.com/en-us/securi...
Reposted by Elliot
Beyond immediate containment, Microsoft IR supports recovery, future planning, and building long-term resilience. According to Adrian Hill, lead investigator for Microsoft IR, “The customer needs to be successful. The only way to do that is to ensure that everyone is successful.”
October 1, 2025 at 7:34 PM
Beyond immediate containment, Microsoft IR supports recovery, future planning, and building long-term resilience. According to Adrian Hill, lead investigator for Microsoft IR, “The customer needs to be successful. The only way to do that is to ensure that everyone is successful.”
Reposted by Elliot
By leading with empathy and collaboration, Microsoft IR unites vendors and internal teams to stabilize crises and uncover hidden threats, ensuring unified action. This approach means that every engagement restores the customer and simultaneously strengthens the broader security ecosystem.
October 1, 2025 at 7:33 PM
By leading with empathy and collaboration, Microsoft IR unites vendors and internal teams to stabilize crises and uncover hidden threats, ensuring unified action. This approach means that every engagement restores the customer and simultaneously strengthens the broader security ecosystem.
Full episode here www.microsoft.com/en-us/securi...
From Insight to Disruption | Security Insider
How Microsoft disrupted Storm-1152’s 750 million fake accounts. See how threat intelligence becomes action, disruption, and protection.
www.microsoft.com
September 17, 2025 at 2:08 PM
Full episode here www.microsoft.com/en-us/securi...
Each episode will offer an inside look at Microsoft Security's threat intelligence capability that is designed to reduce risk, improve resilience, and empower security teams across the globe.
September 17, 2025 at 2:08 PM
Each episode will offer an inside look at Microsoft Security's threat intelligence capability that is designed to reduce risk, improve resilience, and empower security teams across the globe.
This week we are releasing episode one of Inside Microsoft Threat Intelligence, a new series highlighting the power of our 10,000-strong security team.
September 17, 2025 at 2:08 PM
This week we are releasing episode one of Inside Microsoft Threat Intelligence, a new series highlighting the power of our 10,000-strong security team.