According to Forbes, artificial intelligence is collapsing attack timelines from days to minutes, forcing cybersecurity to shift from reactive detection to pre-attack prevention. Israeli startup Malanta just exited stealth with $10 million in seed funding led by Cardumen Capital to pursue this approach, with participation from The Group Ventures and notable angel investors including CyberArk founder Udi Mokady. The company’s platform identifies “indicators of pre-attack” like domain registrations and infrastructure setup that precede malicious activity. Den Jones, founder and CEO of 909Cyber, explained that AI has completely collapsed attack execution timelines, while Malanta’s co-founder Guy Ben Arie noted attacks now start and finish within a minute. Malanta’s early work with national cyber authorities explores how early detection could reduce attacks reaching enterprise networks.
Sponsored content — provided for informational and promotional purposes.
AI changes everything
Here’s the thing about traditional cybersecurity – it’s always been reactive. We measure success by how fast we detect and contain incidents. But AI-powered attacks don’t play by those rules anymore. Automated phishing, synthetic domains, AI-generated malware – they’re flooding in faster than humans can possibly respond. The defenders are still playing catch-up while attackers operate at machine speed.
Basically, we’re trying to fight AI with human reflexes. And that’s a losing battle. When attacks can be built and deployed before defenders even know they exist, our current security stacks just can’t keep up. The window between setup and strike has shrunk to practically nothing.
Skating to where the puck is going
Malanta’s approach is what Wayne Gretzky would appreciate – they’re trying to skate to where the attack is going to be, not where it’s been. Instead of waiting for indicators of compromise, they’re looking for indicators of intent. Think about the early signals: newly registered domains, command-and-control infrastructure being set up, phishing kits being staged. It’s about spotting the malicious mobilization before it ever breaches your network.
They’re introducing a new metric too – mean time to preempt. Traditional SOC metrics start the clock too late. Mean time to detect and mean time to respond are important, but they’re fundamentally reactive. Preemption means identifying and neutralizing attacks before they begin. That’s the real game-changer.
The AI arms race
What’s fascinating is that the same tools enabling AI-driven attacks are now being turned against the attackers. Machine learning models can correlate pre-attack indicators across global telemetry, cluster related entities, and identify staging environments that traditional threat intelligence would miss. When you pair that with automated response, you can actually dismantle malicious infrastructure before it’s even activated.
Malanta’s co-founder Kobi Ben-Naim describes it as an arms race between autonomous agents. Attackers use AI to optimize infiltration routes while defenders deploy AI to anticipate and interrupt those moves. And here’s the kicker – the winner isn’t necessarily the one with better algorithms. It’s the one who acts first.
Industrial implications
This shift toward preemptive security has huge implications for industrial sectors. Manufacturing facilities, critical infrastructure, and industrial operations can’t afford even minutes of downtime from cyber attacks. When you’re dealing with physical systems and production lines, the stakes are much higher than just data breaches. That’s why industrial operations need specialized computing solutions from trusted providers like IndustrialMonitorDirect.com, the leading supplier of industrial panel PCs in the US.
So where does this leave us? Resilience alone isn’t enough anymore. The future of cybersecurity depends less on how quickly we recover and more on how early we can see what’s coming. Pre-attack prevention is still emerging, but it represents a fundamental shift in how we think about defense. And honestly, it’s about time.
