AI found #Java4Shell and helped write the exploits. At this point that’s not news, it’s the baseline. Finding critical vulns is getting cheap.
Mandiant has mean time-to-exploit at -7 days now. Attackers are in before the patch exists. Patching fast still matters, but it’s not a strategy anymore if the exploit beats the fix.
What actually decides how bad it gets is what the compromised asset can reach. Least privilege, segmentation, identity scoping. Boring stuff, done before the incident.
Worth noting: Keycloak is on the affected list. RCE on your IdP is a very bad day if nobody ever scoped what trusts it.
Three breaches, three different attack surfaces:
• DHS: 41K+ employee records exposed, reportedly via SSO.
• Binance: ~$6.3B stolen through abuse of legitimate transaction-signing infrastructure.
• Polymarket: ~$2.7M stolen via compromised third-party JavaScript.
The common thread? Trusted identities, services, and code can become attack paths.
Prevention matters, but so does containment. Security must continuously understand what each identity or service can reach, what actions it can perform, and the potential blast radius if compromised.
The goal shouldn't be only stopping the initial breach. Breaches will happen. t's to ensure that one compromised component doesn't become a systemic failure.
Binance's initial investigation suggests that the security incident may involve a compromised DevOps engineer’s laptop, which allowed an attacker to gain access to a service container and submit unauthorized withdrawal transactions. We have found no evidence of any wallet private key leak. We will share a detailed technical post-mortem once the investigation is complete and the facts are fully verified.
The estimated total impact sits at around $6.3B. We have worked with blockchain partners to freeze affected assets, with ~$3 billion frozen on-chain so far.
Binance will cover all losses from the SAFU fund and the company’s funds. User funds are safe. We will also publish an updated Proof of Reserves once the remediation is complete.
As an immediate precaution, Binance has:
- temporarily suspended deposits and withdrawals while security checks are completed;
- revoked the affected address authorisation;
- engaged law enforcement and blockchain intelligence partners to trace and recover funds.
Our teams are working around the clock to resolve this issue safely and thoroughly. We are working closely with our law enforcement partners to confirm the identity of these criminals and bring them to justice.
Please rely only on official Binance channels for updates. Please feel free to contact Binance Customer Support if you have any further questions.
https://t.co/ywB2mKFy0p
I agree with the urgency, but I think "Patch or Die" misses an important point.
You can't patch a zero-day before it's discovered, and you can't patch everything immediately.
The more sustainable defense is resilience: continuously eliminating unnecessary exposure, privileges and attack paths, so that even when something gets exploited, the damage is contained.
Patch fast, absolutely. But design your environment to survive what you haven't patched yet.
Patch or Die!
Today we sent the following out to the CEOs in our portfolio regarding the widespread vulnerabilities that have been detected by AI agents:
The previous week should make it obvious that things have changed. Vulnerabilities are being discovered and exploited at a rate that has never been seen before. While there has been a flood of new vulnerabilities AI agents are being adopted at scale multiplying the potential risk. Agents not only help automate the attacker lifecycle, every ai agent is an insider threat. Your environments are designed to withstand attacks from the outside, not from within. As such, we are suggesting company leaders consider the following changes:
1 - Ensure you're sufficiently staffed for security. The average enterprise allocates roughly 10% of its tech spend to security. Consider doubling that. There are likely thousands of vulnerabilities that need fixing in your environments. Security through obscurity no longer works, all that can be exploited will be exploited.
2 - Establish a zero tolerance policy for secuirty bugs in new features and product releases. Re-examine launch and release schedules and consider delays to ensure all vulnerabilities are patched before deploying new features or products.
3 - Double tap the zombies. Find all old legacy environments that have been neglected and turn them off. Labs, test environments, zombie CI/CD projects etc. Turn it off for now. If it isn’t tied to revenue, kill it.
4 - End the risk acceptance process. Over the last 10 years the risk acceptance process has morphed from a way to accept risks to enable the business, to a way to justify never fixing security issues. Adopt a zero tolerance policy for security issues and risks.
AI isn't just making attacks more sophisticated.
It's changing their economics.
Work that once required a skilled team for days or weeks can increasingly be done continuously and in parallel by agents.
That changes defense too. Counting patches is not enough. We need to understand which attack paths are actually viable, and where to break them.
I discussed this, context graphs, agentic exposure management and the future of security operations with James Kaplan:
https://t.co/f2SvA99i80
The threat model for defenders is changing.
The cyber capability gap between frontier and open models is closing faster and faster, while many of the safeguards are simply gone.
This will only get harder to defend against.
https://t.co/gkxb6Rytml
https://t.co/7stWhJsLpH
What's exciting about Jev isn't speed/cost, it's the shape: frontier judgment as a function call, not a conversation. That pulls the work up a level. You stop engineering prompts and start designing the decision structure - which calls, at what granularity, with what thresholds.
Anthropic’s latest threat report highlights a shift I think is more important than “AI makes attackers better”:
AI is changing the economics of attacks.
When agents can continuously perform reconnaissance, vulnerability research and exploitation, many attack paths that weren’t worth a human attacker’s time suddenly become viable.
The defensive question changes too: not just “how severe is this vulnerability?” but “what paths can an autonomous attacker actually complete, and where can we break them most effectively?”
https://t.co/5QYjRqy9Np