Source Transparency
Microsoft Research Blog
Recent VQV signals collected from this public source, grouped with the topics where it appears.
Source health details are not available for this source yet. Recent signal count and last seen time are shown from public findings.
Recent Signals
All sources
USEFUL NOW · SOURCE-BACKED
95% signal strength
Microsoft researchers describe Lazarus, a backdoor that reprograms dormant neurons in AI agents
Microsoft researchers have described a backdoor technique called Lazarus that can reprogram dormant neurons in an AI model. According to the source snippet, the technique can cause an otherwise normal AI agent to execute malicious commands when a hidden trigger appears. The snippet notes the threat is especially serious for agents with tool or terminal access, where a triggered model could quietly download and run attacker code. The snippet is excerpted, so further details of the research are not included here.
Why it matters: Developers and businesses deploying AI agents with tool or terminal access may need to consider backdoors that stay hidden until a trigger appears. Security teams evaluating agent deployments should watch for follow-up technical details and any mitigations Microsoft publishes.
Why this is here: This signal is recent, source-backed, and connected to activity readers are already following in AI Agents.
RESEARCH · SOURCE-BACKED
92% signal strength
Microsoft Research Focus Edition #97 covers coding agents, AI at work and weather forecasting
Microsoft Research published Edition #97 of its Research Focus podcast, hosted by Sowbhagya Suresh. According to the episode description, it shares recent news and research from across Microsoft, including the challenges facing coding agents, how people use AI at work, and improving extended-range weather forecasting. The snippet does not give specific findings, figures or dates, so the details of each topic are not available from this source alone.
Why it matters: For developers and teams building with coding agents, the episode points to ongoing Microsoft research into the difficulties these tools face. People studying workplace AI use may find the research on how employees actually use AI relevant to their own work. Readers should check the full episode or linked research for concrete results before drawing conclusions.
Why this is here: This signal is recent, source-backed, and connected to activity readers are already following in AI at Work.