Part 6: The playbook
Part 6 of 6 in a deep dive on machine learning that predicts attacks and stops them early, with Apple as the main case. Start with Part 1 if you are new to the series.
Five parts followed one attack inward, from the device to the exploit, the locks and the physical door. Attackers adapt, locks can be fooled, and locks can jam on their own. This last part asks what a defender should build in response, and borrows the answer from the oldest defense system there is.
What evolution teaches
Nature does not answer adaptive attackers with one perfect lock. The immune system keeps a varied population of detectors and keeps replacing them, and Taleb's 1/N rule says the same in statistical terms: keep each error small and separate, so no single one spreads through everything.
My open-source library genetic_xgb applies that loop to XGBoost, the gradient-boosted trees that suit the tabular data every lock produces, including the physical fingerprint. It keeps a population of candidate models, scores each on data it was not trained on, keeps the best, breeds and mutates them, and stops when the score stops improving. A population of models that rely on different signals is also harder to fool with one forged badge or one cloned voice.
Five ways to use it at the physical door
- Evolve the fingerprint model
- Train on badge, camera, phone, Wi-Fi and door-sensor features, with fitness set to catching borrowed badges at a fixed rate of false holds.
- A pickup verification score
- Score each pickup or vault release on email domain age, phone history, ID scan, plate and timing, and evolve the model on confirmed fictitious pickups.
- A false-alarm filter per site
- Each site gets its own population, trained on its own reviewed alarms, because a warehouse at night and a lobby at noon look nothing alike.
- Drone versus bird, refreshed
- Re-evolve the fusion classifier as new drone types and seasons change the radar and sound signatures.
- Shadow mode first
- Run every evolved model alongside the current one for a month before it opens or holds a single door.
The signal in the 2026 noise
There were hundreds of security reports in 2026. Six findings held up across the independent sources in this series, and they are the ones I would build next year's plan around.
- Exploits are now the main way in
- Vulnerability exploitation passed stolen credentials as the top entry point in Verizon's data, at 31% (Verizon), and Mandiant puts exploitation at seven days before the patch on average (Mandiant).
- Attackers use AI for scale, not magic
- Verizon found threat actors using AI across about 15 attack techniques, with less than 2.5% of them new (SOCRadar on Verizon). JADEPUFFER showed the next step, a whole attack with no person at the keyboard (GCN).
- Speed decides
- Average breakout fell to 29 minutes (CrowdStrike), while a triage agent handles an alert in about a minute (TechInformed).
- Identity is the weak point, online and at the dock
- Phishing, social engineering and valid accounts made up 73.3% of initial access attempts (Microsoft), and fictitious pickups rose to about a quarter of cargo incidents (FleetOwner).
- AI is now part of the attack surface
- AI models or apps were involved in 21% of breaches, and 92% of those had no proper AI access controls (Help Net Security).
- Correlation beats any single sensor
- Microsoft calls signal correlation the primary strategic variable (Microsoft). Fused sensors at an airport and a physical fingerprint at a vault are the same idea.
Final thoughts
This series followed an attack inward. Apple locks the device in silicon. Exploit prediction tells defenders which bug will be used before it is. Models at the edge, the sign-in page and the help desk decide in seconds, and can themselves be attacked or fail. At the physical door, the same AI forges the identity that opens a loading dock, and the same AI, fused across cameras, radar, Wi-Fi and badge readers, can tell the borrowed badge from the real one.
What the models do
They predict which door will be tried, score every sign-in, request, frame and radar return in seconds, and combine them into one profile that is hard to fake.
What evolution adds
Randomness that keeps attackers from learning the system, a population of models that fail one at a time, staged rollouts and honest labels from reviewed alarms.
The cheapest place to start is the release decision, the moment a load, a box or a bar leaves your control. Log every release with its email, phone, ID and timing signals, label the ones that turn out to be fraud, and evolve a small population of models to score new releases in shadow mode for a month. Then put the best one in front of the person who signs the release, before it holds anything.
Sources
- M-Trends 2026: data, insights and strategies from the frontlines, Mandiant, Google Cloud
- Cargo theft losses rise 39% as fictitious pickups remain a threat, FleetOwner, November 2026
- Cargo crime report, The Loadstar, October 2026
- Chaucer backs Lloyd's cargo program targeting strategic theft, Insurance Business, October 2026
- Cargo theft report Q1 2026, Insurance Journal, April 2026
- The year the rich went wild for gold, Financial Times, August 2026
- Demand for private vaults soars in London, Spear's
- Louvre thieves escaped with 30 seconds to spare, Senate inquiry, December 2025
- One year after Louvre heist, AFP via Kuwait Times, October 2026
- IEEE Std 802.11bf-2025, WLAN sensing amendment
- IEEE 802.11bf: enabling the widespread adoption of Wi-Fi sensing, NIST
- New IEEE Wi-Fi sensing standard and WhoFi research, overview
- Motorola Solutions video surveillance strategy, Memoori, 2026
- Hikvision 2025 full-year results, SecurityWorldMarket, May 2026
- ASSA ABLOY Q4 2025 earnings summary, Quartr
- Securitas full year report 2025, Inderes
- Securitas Q4 2025 presentation
- SECOM company overview FY2026, Envalith
- SECOM Q1 FY2026 earnings summary, Quartr
- Hanwha Vision 2025 results, Hellot
- Hanwha Vision record 2025 results, Boannews, February 2026
- dormakaba FY 2025/26 results, wallstreet-online
- dormakaba FY 2025/26 press release, finanzen.at
- Brink's Q4 2025 earnings summary, Quartr
- Loomis full-year report 2025, Inderes
- Security industry overview 2025, National Police Agency of Japan
- 2025 information security industry survey, Ministry of Science and ICT via Boannews
- Physical security market 2026–2034, Fortune Business Insights via GII
- Private security services in Switzerland, IBISWorld
- Report on private security personnel in Switzerland, Swiss federal government
- Microsoft Digital Defense Report 2026, executive summary, October 2026
- Microsoft highlights rising cyber risks, Security MEA, October 2026
- What the Microsoft Digital Defense Report 2026 means for sign-in, WorkOS
- Microsoft report documents first fully autonomous ransomware attack, GCN, October 2026
- JADEPUFFER-linked attackers used compromised service principals in Azure, The Hacker News, September 2026
- JadePuffer uses AI agents to wipe Azure resources in minutes, Windows Report
- JADEPUFFER agentic ransomware, Decryption Digest
- The State of Physical Security Operations in 2026, HiveWatch, June 2026
- Eyeforce and ABC Supply Co case study, Pelco
- Drone disruptions at Copenhagen and Munich airports, Sentrycs
- DIDIT response to rising airport drone threats, Telespazio
- The drone blockade: airports grapple with a growing threat, Defence Online, November 2025
- Drone detection at airports, a&s, 2026
- Danish and UK firms rush counter-drone takeover amid airspace breaches, C4ISRNET, December 2025
- Germany airport drone disruption management market, MarkNtel Advisors, 2026
- Drone threats drive surge in airport defence tech, Reuters via Sowetan, June 2026
- Vulnerability exploitation top breach entry point, 2026 DBIR, Verizon, May 19, 2026
- 2026 Verizon DBIR: the new era of cyber threats, eSecurity Planet
- New Verizon report reveals the security gap attackers are exploiting most, TechRepublic, May 2026
- Verizon 2026 DBIR: 10 takeaways, SOCRadar
- 62% of breaches involved the human element: key takeaways from Verizon 2026 DBIR, Abnormal
- IBM Cost of a Data Breach 2026, Help Net Security, July 30, 2026
- IBM 2026 Cost of a Data Breach report: key findings, eSecurity Planet
- Shadow AI data breach cost 2026, SureCloud
- Shadow AI doubles to 43% of breaches, Forkast
- Google's SOC agent cuts alert review from 30 minutes to 60 seconds, TechInformed, June 2026
- Microsoft Security Copilot agents, Microsoft
- CrowdStrike 2026 Global Threat Report, CrowdStrike
- AI now routine in cyber attacks, Google report finds, IT Brief, May 2026
- Cloudflare outage on November 18, 2025, Cloudflare
- Cloudflare blames Bot Management system for widespread outage, CyberInsider
- The file that paralyzed 20% of the Internet, EveryWAN
- Google report on adversarial AI use, SiliconANGLE, February 12, 2026
- Report: adversarial use of AI is evolving, KnowBe4 on GTIG, May 2026
- Google: AI now powers every threat actor's playbook, Cyber Magazine on GTIG's Q3 2026 tracker
- CrowdStrike 2026 Global Threat Report findings, CrowdStrike, February 24, 2026
- 2026 CrowdStrike Global Threat Report press release, via 01net
- How faster cyber-attacks are reshaping cybersecurity strategy, Infosecurity Magazine
- CrowdStrike 2023 Global Threat Report, MSSP Alert
- Announcing M-Trends 2026, Google Cloud Security Community
- FBI report adds AI-enabled scams for the first time, FOX, April 2026
- FBI releases annual internet crime complaint report, April 2026
- Deepfake statistics 2026, Keepnet, citing Gartner
- Anthropic's Project Glasswing finds 10,000+ CVEs in one month, Daily Security Review
- AI vulnerability discovery velocity and the disclosure crisis, Cloud Security Alliance, May 2026
- Claude Mythos turns years of security research into 20-hour AI exploits, TechRadar
- Threat actors are exploiting vulnerabilities faster than ever, TechRepublic on Mandiant's 2023 analysis
- Tenable CTO Q&A on AI threats and exposure management, Tenable
- The vulnerability curve bent with the AI curve, ProjectDiscovery
- Zero-day exploit statistics, DeepStrike, citing Google Threat Intelligence Group
- Attackers are handing off access in 22 seconds, Help Net Security
- CVSS vs EPSS vs SSVC vs KEV, DefenderMate
- Introducing EPSS version 4, Jay Jacobs, Empirical Security, March 17, 2025
- Improving vulnerability remediation through better exploit prediction, Jacobs et al., 2019
- What is EPSS, SecPod
- Google AI Big Sleep stops exploitation of SQLite flaw, The Hacker News, July 2025
- Google's new AI doesn't just find vulnerabilities, it rewrites code to patch them, The Hacker News, October 2025
- Google DeepMind unveils CodeMender, Ground News coverage summary
- Memory Integrity Enforcement, Apple Security Research, September 9, 2025
- Operating system integrity, Apple Platform Security
- BLASTPASS: NSO Group iPhone zero-click, zero-day exploit captured in the wild, Citizen Lab, September 7, 2023
- Apple says it is not aware anyone using Lockdown Mode got hacked, TechCrunch, December 2023
- Apple doubles maximum bug bounty to $2M for zero-click RCEs, Security Affairs, October 2025
- Security research on Private Cloud Compute, Apple Security Research, October 2024
- About Apple threat notifications and protecting against mercenary spyware, Apple Support
- Apple warns users in 110 countries of mercenary spyware, TechRepublic, August 2026
- Apple notifications alert victims in 99 countries, CyberScoop, October 2026
- Apple helps safeguard users from scams and fraud while protecting personal data, Apple Newsroom, October 2025
- About Stolen Device Protection for iPhone, Apple Support
- iPhone security delay: all you need to know, NordVPN
- How Cloudflare auto-mitigated a world record 3.8 Tbps DDoS attack, Cloudflare
- Cloudflare thwarts record 7.3 Tbps DDoS attack with automation, SecurityBrief
- Cloudflare auto-mitigates 11.5 Tbps attack, SecureBlink
- Record-breaking DDoS attack peaks at 22 Tbps and 10 Bpps, SecurityWeek
- Cloudflare report shows massive spike in high-volume DDoS attacks, TechRepublic
- DDoS attacks over 1 Tbps surged fivefold in the second quarter, BleepingComputer
- DDoS attacks hit record scale as 1 Tbps+ campaigns become more common, Help Net Security
- Cloudflare reports 18.2 million network-layer DDoS attacks in Q3, The Hacker News
- Cloudflare mitigates 23.2 million DDoS attacks, GBHackers
- Microsoft Digital Defense Report 2025 summary, Microsoft
- Microsoft Digital Defense Report 2025, Libertify summary
- Google deploys on-device AI to thwart scams on Chrome and Android, Infosecurity Magazine
- How we're using AI to combat the latest scams, Google
- Keeping Google Play and Android app ecosystems safe in 2025, Google
- Google says its AI systems helped deter Play Store malware in 2025, Reuters via Business Ghana
- Train, evaluate, repeat, Michael Manapat, PyData LA 2018
- Counterfactual evaluation of machine learning models, Michael Manapat (Stripe), slides
- Adversarial machine learning: a taxonomy and terminology of attacks and mitigations, NIST AI 100-2e2025
- CrowdStrike shares how a Rapid Response Content update caused global outage, Infosecurity Magazine
- CrowdStrike blames outage on content configuration update, Computer Weekly
- Disrupting the first reported AI-orchestrated cyber espionage campaign, Anthropic, November 2025
- Anthropic disrupts AI-run cyberattack, Decrypt, November 2025
- Leigh Van Valen, evolutionary theorist and paleobiology pioneer, University of Chicago News
- Self-nonself discrimination in a computer, Forrest, Perelson, Allen and Cherukuri, 1994
- genetic_xgb, aleator-labs on GitHub
- Hardware wallets, randomness and the vaults behind every device, Zenpanik
- Two clocks: quantum breaks the math, AI breaks everything around it, Zenpanik
- Nassim Nicholas Taleb, letter to Stewart Brand on fragility and precaution, July 3, 2013