Why SIGNALS
Every year the security industry publishes hundreds of reports. Most are careful work, and each one makes the case that its problem is the biggest. For a CEO or a board, the result is noise.
Over the past two years, FIR Risk has published 100 editions analyzing the major reports, from Verizon, Microsoft, Mandiant, CrowdStrike, IBM, Anthropic and many others. This year we asked a different question: what survives when you put them all side by side?
SIGNALS is the answer. A finding made the cut only if independent sources agree on it, it has held for more than a year, and it matters to the decisions leaders make about priorities, funding and oversight. We went back to the original reports and checked all 43 figures. What didn’t pass, we left out.
Thank you for reading FIR Risk this year.
— Bruce, FIR Risk Advisory
Bottom Line
Attackers now move faster than any patch cycle, look like your own employees once they’re inside, and increasingly come in through your vendors. Fewer victims pay ransoms, so the cost of an attack is now set by how long you’re down. And the AI risk you can actually control is your own company’s use of it. The good news: the basics still stop most attacks, and companies are catching more of them on their own.
Five signals, five questions. None of them takes long to ask.
The One Sentence Your Board Needs
Which basics are we not funding yet?
Ask it before the other five. More than 90% of breaches were enabled by gaps that were preventable, and multi-factor authentication still blocks more than 99% of unauthorized sign-ins. The fundamentals still decide most outcomes.
The five signals
01 · The window is gone. Expose less.
Average eCrime breakout time, from first access to moving through the network, fell from 84 minutes in 2022 to 29 minutes in 2025. The fastest was 27 seconds. And 29% of exploited flaws were attacked on or before the day they were published. You can’t out-patch that. You can shrink what you expose.
Ask: Do we know every system we expose to the internet and every open-source package we depend on, and is that footprint getting smaller?
02 · Inside, they look like you.
Logins are harder to break, so attackers talk their way in. Voice phishing is now the #2 way in. Once inside, they act like real users: 82% of the intrusions CrowdStrike detected in 2025 used no malware at all, just stolen logins and the tools your own staff use.
Ask: If an attacker were logged in as one of our employees today, would we know?
03 · Half of breaches now involve a third party.
48% of breaches involved a vendor, partner or supplier, up from 9% three editions ago. In one case, a single stolen token reached full control of a company’s cloud in about three hours. Trusted relationships were the way in for 20% of extreme cyber events, but 65% of their losses.
Ask: How many vendors do we rely on, and how many of the critical ones have a current risk assessment?
04 · Recovery time sets the bill.
69% of ransomware victims refused to pay, up from about half in 2022. Yet data-extortion events rose 63%, and attackers now go after your backups and identity systems. Loss grows much faster than downtime: in one economic model, taking almost two weeks to restore cost about 10× as much as restoring within one.
Ask: How quickly could we really restore core operations, and have we tested it?
05 · Your own AI is the risk you control.
Fewer than 1% of the threat actors caught using AI were rated high or critical risk: AI made attacks cheaper and faster, not more successful. Inside companies, AI use tripled, from 15% to 45% of employees in a year, and 67% of AI users sign in with personal accounts. Unsanctioned AI added about $670,000 to a breach’s cost.
Ask: Which AI tools and agents are in use across our company, approved or not, and who owns each one?
What else is in the report
- The good news. MFA blocks more than 99% of unauthorized sign-ins. The share of breaches organizations found themselves rose from 43% to 52% in a year.
- The noise we filtered out. Six loud claims that failed our tests, from “AI super-attacks are here” to “a breach costs $4.44M.”
- How our past calls held up. We graded the threats named in our 2024 and 2025 reports: 4 held, 1 changed shape, 1 partly right.
- Weak signals to watch in 2027. Attacks run by AI agents, AI on the defender’s side, post-quantum deadlines, intruders staying unseen longer, and prompt injection.
- Every number, checked. All 43 figures, each traced to the page of the primary report it comes from.
- For your security team. A two-page appendix mapping each signal to MITRE ATT&CK and ATLAS techniques and mitigations.
Read it, watch it, or listen
- Watch: the narrated video, with the report’s charts and figures on screen (10 minutes, captions included)
- Listen: the narrated edition (10 minutes)
- Read: the 18-page PDF for your board pack, or the full web edition
All free, no registration: firrisk.ai/reports/signals-2026
Learn More
- FIR SIGNALS 2026 — the full report, video, audio and PDF
- FIR Risk Tuesday E97 — The Capacity Race — what big tech’s own filings say about technology risk
- FIR Risk Tuesday E96 — The Costliest Five Percent — the extreme-event data behind signals three and four
- How we verify — how every number is checked at the source
SIGNALS is general risk intelligence for leaders. It is not legal, compliance or security advice for a specific organization. FIR Risk uses AI models from several providers, including Anthropic.
