Adversarial AI is attacking validator infrastructure. We defend it with adversarial AI - trained on the network itself, acting at machine speed, only when authority has been earned. Zero latency impact.
Every event above is counterfactual - what IBSR would have done under earned authority. Live enforcement begins only on your consent, scoped to a class of abuse, revocable at any time.
AI agents now discover and chain zero-day exploits autonomously, at $50 a vulnerability. Validator infrastructure - trillions in staked capital, hundreds of independent operators, no central SOC - is uniquely exposed.
A DoS-plus-consensus exploit on Litecoin's MWEB caused a 13-block reorganisation with roughly $600K in cross-chain exposure. The patch had been available for 37 days. Not a zero-day - a known-day that defenders couldn't close in time.
Two major decentralized networks were hit with sustained DDoS in the same week. Same wave, different outcomes. The difference was the defensive substrate underneath - most chains don't have one.
One brain. Two open-source nodes. The network defends itself - at kernel speed, with authority earned, not asserted.
Collective intelligence, trained on the network it defends.
Observes, learns, reports home - on your traffic, not synthetic data.
Kernel-speed action when authority has been granted.
There is no MITRE ATT&CK for validator infrastructure. No CVE pipeline. No shared format for what an attack on a decentralized network actually looks like. The industry has no substrate.
We built one. An open multi-modal bundle format, an open ten-family taxonomy, and a proprietary corpus of 1,092 contract-validated attack bundles across 19 primitives.
Read the research →Most autonomous defence asks for blind trust. We don't. IBSR runs in shadow mode on your traffic, producing counterfactual evidence of what it would have done. You decide when the evidence is sufficient.
Authority is granted in stages, scoped to specific abuse classes, and revocable at any time. The framework is published, peer-reviewable, and open.
Read the framework →IBSR observes packets in passive mode. No enforcement. Counterfactual evidence accumulates.
Operator grants authority for one abuse class. Guard begins enforcing for that class only.
Authority expands by evidence, never by request. Each class is opt-in and revocable.
Cohort-wide policy. One validator's judgment becomes another's protection.
8 founding seats. We are assembling the first cohort now - operators who have felt the wave and want a hand on the rudder while the substrate is still being written.
Right now we need your time and your telemetry - not your trust. Later: full platform access, roadmap influence, and founding-cohort pricing locked in.
You run validators on production networks and feel the wave first. You want telemetry on what is actually probing your nodes.
You serve the network's reads and absorb its noise. DDoS, scraping and RPC abuse are your daily reality.
You set protocol policy and own the defensive substrate. You can authorise a cohort and shape the taxonomy.
You run nodes at scale for others. One judgment on your fleet protects every delegator behind it.