Gremlin launches AI tool to speed up infrastructure resilience testing
Gremlin, the chaos engineering startup founded by a Netflix engineer, has launched Foresight AI, an AI-powered addition to its system resilience testing services. The new tool automatically breaks software infrastructure to uncover hidden bugs before they reach production, whilst also diagnosing problems and proposing fixes. This automation significantly speeds up the chaos testing process, which traditionally required extensive manual preparation and analysis.
The AI is powered by Gremlin's "Failure Atlas," a repository containing millions of chaos engineering experiments conducted over a decade on tens of thousands of systems. Rather than relying solely on general knowledge, the system grounds its recommendations in this specialised database of real-world failures, minimising hallucinations and providing technically informed advice. Once Foresight AI identifies a failure's root cause, it can generate code or configuration changes to fix the issue, either applying them automatically or presenting recommendations for human review.
- Gremlin launches AI to automate chaos testing and detect hidden bugs faster
- System uses a decade of data from millions of experiments for accurate diagnoses
- AI can autonomously identify root causes and propose fixes for distributed system failures
New here? Start with this
Companies increasingly rely on complex software systems that must work reliably. Before software reaches customers, engineers deliberately break these systems to find hidden weaknesses in a practice called chaos engineering or resilience testing. This helps them fix problems before they cause real outages that might lose money or upset users.
Gremlin is a company that helps organisations conduct this type of testing. It was founded by an engineer who previously worked at Netflix, the streaming giant that pioneered many chaos testing techniques. The company provides tools to help businesses identify and fix problems in their critical systems.
Traditionally, infrastructure testing required considerable manual work, with engineers having to carefully design experiments and spend significant time analysing results. Automating more of this process matters because it makes thorough testing faster and more practical for organisations without large engineering teams, potentially preventing outages that affect customers.
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Automating chaos engineering through Foresight AI represents a genuine advancement for infrastructure resilience because it accelerates the identification and remediation of vulnerabilities whilst being grounded in Gremlin's specialised, real-world Failure Atlas rather than relying on general-purpose AI knowledge. By dramatically reducing the time and manual effort required for resilience testing, organisations can detect and fix bugs more quickly, preventing production failures and improving system stability. The tool maintains the option for human review, allowing teams to retain meaningful oversight whilst benefiting from AI-assisted analysis and automated fix recommendations.
The case against
Automating critical infrastructure testing and repairs raises legitimate concerns about concentrating decision-making in a single system and eroding human expertise in mission-critical processes. Although Gremlin's Failure Atlas is extensive and grounded in real-world data, it necessarily reflects patterns from past failures and may not identify novel or unprecedented failure modes that experienced engineers would catch through traditional testing. The ability to automatically apply infrastructure changes without mandatory human review introduces risks, as even well-designed AI systems can make confident errors with cascading consequences in complex, interdependent systems.
Read the full article at the source →
Originally published by The Register as “Gremlin now uses AI to break distributed systems faster”.