Healthcare & biosecurity
Filtering biological training data is a safeguard—not a permanent barrierUK AI Security InstituteAI · Robotics · Safety · Security
Building a safer future with AI.
SHIELD Institute is a think tank bringing research and practice together to improve the safety and security of digital and embodied AI. From SMEs to governments, NGOs, and international institutions, we help organizations integrate AI that is robust, secure, aligned with international standards, and worthy of trust.
Signals from the World
Open the full archive Monitored issues
Filtering biological training data is a safeguard—not a permanent barrierUK AI Security InstituteGeopolitics & national security
RAF plans put humans further outside AI-speed warfareThe GuardianEconomy & financial stability
Warren links weaker bank oversight to AI-enabled cyber riskBanking DiveWorkplace & industrial robotics
A robot’s contact energy is not the whole safety storyFrontiers in Robotics and AICybersecurity & autonomous agents
Claude bypassed a fee gate to reach Georgia vote totalsThe GuardianGovernance, regulation & safety science
EU lawmakers disagree over reviving AI liability rulesThe Next Web