Israel startup launches real-time AI agent blocking
Tap title to read story
Sweet Security Ltd., an Israeli runtime security startup, has introduced Agentic AI Blocking, a new capability designed to halt artificial intelligence agents mid-action when they attempt unauthorized tasks in production environments. Announced ahead of Black Hat USA 2026 in Las Vegas, the technology extends runtime security from traditional cloud workloads to autonomous agents operating alongside them. The system can instantly terminate unverified tool calls and active agent sessions, preventing the exfiltration of sensitive corporate data, secrets and personally identifiable information. It also detects and shuts down prompt injection attempts before malicious instructions can alter an agent’s intended behavior inside enterprise environments.
More context
Timeline
- On January 19, 2025, former president Joe Biden issued preemptive executive pardon.
- On May 1, 2026, Sweet Security brand announced automated runtime security platform developments.
- On July 29, 2026, Sweet Security debuted agentic AI blocking at Black Hat.
- On July 29, 2026, tech startup revealed active runtime threat prevention capabilities.
- On July 29, 2026, industry reports highlighted rising automated agent security vulnerability risks.
- On July 29, 2026, cybersecurity experts analyzed enterprise threat mitigation agent blocking protocols.
- On August 1, 2026, organizations will integrate advanced agentic security blocking frameworks.
- On September 15, 2026, enterprises will deploy runtime defenses against rogue agent attacks.
- On December 31, 2026, corporate security standards will adapt to autonomous agent threats.
- On July 29, 2027, automated agent runtime security will become standard enterprise software.
Why it matters
New runtime security stops unauthorized enterprise artificial intelligence agent actions. Agentic threat blocking safeguards corporate data against autonomous system exploits. Enterprise IT security teams, cloud administrators, and software development corporations. Monitor enterprise cybersecurity advisories and emerging runtime threat mitigation reports.
Sources
JQJO Editorial Desk · Human reviewed · AI assisted. Report a correction.
