📊 Full opportunity report: The Frameworks Can’t See the Thing That Matters: A Year of AI-Enabled Cyber Threats on ThorstenMeyerAI.com — validation score, market gap, and execution plan.
TL;DR
Anthropic’s analysis of a year of malicious AI use shows attackers increasingly leverage AI for complex tasks, challenging existing threat evaluation methods. The gap between skill and threat level is narrowing, raising new security concerns.
A new analysis from Anthropic reveals that AI is significantly enhancing the capabilities of cyber attackers, with malicious actors increasingly using AI for complex operations inside networks, challenging existing threat assessment methods.
Anthropic examined 832 accounts banned for malicious activity between March 2025 and March 2026, mapping their techniques onto MITRE ATT&CK. The findings show that 67.3% of these actors used AI primarily to prepare for attacks, such as malware creation, with a notable shift towards using AI for post-breach activities like lateral movement and account discovery. Learn more about how AI impacts cyber threat frameworks.
The proportion of actors employing AI for high-risk, complex tasks increased from 33% in the first half of the year to 56% in the second half, indicating a rapid escalation. Importantly, attackers’ use of AI moved away from initial access techniques like phishing toward deeper network navigation, making the threat more insidious. The report emphasizes that AI now enables less skilled actors to perform activities previously requiring technical expertise, thus democratizing attack capabilities and increasing overall threat levels.
The frameworks can’t see the thing that matters
For decades, danger meant which techniques an attacker commands. A year of real AI-enabled attacks — 832 banned accounts mapped onto MITRE ATT&CK — shows that signal breaking, just as a new, harder-to-see one takes over.
A year of real misuse, mapped to the standard taxonomy
A window, not a census — these are the cases with enough detail to assess techniques thoroughly. Inside it, the risk level climbed fast.
WHAT WAS STUDIED
THE RISK CLIMB · MEDIUM-OR-HIGHER ACTORS
AI-POWERED CYBERSECURITY OPERATIONS: Threat intelligence anomaly detection and automated incident response systems
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
“More techniques” stopped meaning “more dangerous”
The old heuristic: count the techniques, judge the tooling. AI dissolved it — because the model supplies the techniques either way. Watch the old signal fail, then watch what it misses.
Risk score vs. technique count
Two ways to read the same attacker. One is going blind. Press play.
Artificial Intelligence for Cybersecurity: Develop AI approaches to solve cybersecurity problems in your organization
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Deeper into the attack — and into less-skilled hands
Across the year, AI use drifted from getting in toward acting once already inside — the operationally demanding stages that used to require an expert.
The attack lifecycle · where AI is now applied
The center of gravity moved right — toward post-compromise work.
The Practice of Network Security Monitoring: Understanding Incident Detection and Response
Used Book in Good Condition
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
From “what they know” to “what they’ve built”
The report sorts the signals into three tiers — one dead, one fading, one durable.
Technique count & tooling
16 vs. 20 between novice and expert; platform doesn’t correlate. The model supplies the techniques either way.
Where in the lifecycle AI is applied
Concentrating on operationally demanding, post-compromise stages is a better signal — but it’s eroding as the whole population heads there.
The scaffolding around the model
Architectures that let the model chain stages and run with minimal human input. Not what they know — whether they’ve built a system that lets AI run the attack.
Malware Analysis and Detection Engineering: A Comprehensive Approach to Detect and Analyze Modern Malware
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Fixing the map before the territory moves again
A taxonomy that can’t name the most dangerous behavior on the field will quietly mislead the people relying on it. The response runs in two directions.
Fed back into the models
The findings informed safeguards on the most capable models, built to detect & block some of what was observed:
- Blocking malware development
- Blocking mass data exfiltration
- Putting tools in defenders’ hands first (Project Glasswing)
Taking it to the source
Following the Verizon work, Anthropic says it’s in discussions with MITRE about how ATT&CK might evolve:
- A vocabulary for agentic orchestration
- Naming the scaffolding that makes a model an operator
- An interactive technique visualization on the Red blog
Reading it in proportion
- The 832 cases are a detailed subset, not the full population — the precise percentages are directional, not definitive.
- “More autonomous” is not “fully autonomous” — even the standout case needed human input at key moments, which is itself a place for defenders to intervene.
- This is one vendor’s window — the company with visibility into misuse of its own model, publishing what it found. The right thing to do with the data, and worth remembering as you read it.
Implications of AI-Driven Attack Capabilities
This development fundamentally alters threat assessment paradigms. The traditional metric—number of techniques or tools used—no longer reliably indicates threat level, as AI can supply or automate techniques regardless of attacker skill. The shift toward deeper, post-breach activities suggests a more dangerous landscape where less skilled actors can cause significant damage, potentially overwhelming security defenses and complicating attribution. See how AI is changing threat assessment.
Evolution of Cyberattack Techniques and AI Integration
For decades, cybersecurity relied on threat actors’ technical skill and variety of techniques to gauge danger. The MITRE ATT&CK framework has been a standard for classifying tactics and techniques. Recent advances in AI, however, have begun to blur these lines, enabling even less experienced actors to perform sophisticated operations. The analysis from Anthropic builds on prior concerns about AI’s role in cybercrime, highlighting a sharp increase in AI-assisted malicious activity over the past year.
“The traditional heuristic of threat assessment—technique count and tooling—no longer reliably indicates danger in an AI-augmented environment.”
— Anthropic report authors
Unclear Impact of AI on Threat Detection Metrics
It remains uncertain how security frameworks will adapt to these changes or whether new metrics will emerge to better assess threat levels in an AI-augmented landscape. The long-term effectiveness of current detection methods in identifying AI-enabled attacks is still being evaluated, and the full scope of AI’s role in future cyber threats is not yet clear.
Next Steps for Cybersecurity in an AI-Driven Era
Security professionals are expected to reassess threat models and develop new detection strategies that account for AI’s role in attack complexity. Ongoing research will focus on identifying reliable indicators of threat severity beyond technique count, and on creating adaptive defenses capable of countering AI-augmented adversaries. Discover how cybersecurity strategies are evolving with AI.
Key Questions
How does AI change the way attackers operate?
AI enables attackers to automate complex tasks such as lateral movement and account discovery, which previously required significant technical skill, thereby lowering the barrier to executing sophisticated attacks.
Does this mean traditional threat assessments are no longer valid?
Yes, the report suggests that metrics like technique count and tooling are less reliable indicators of threat level in an AI-augmented environment, prompting a need for new assessment methods.
What can organizations do to defend against AI-enabled attacks?
Organizations should update their threat detection strategies to focus on behavioral indicators and operational signals, and invest in adaptive security systems capable of responding to AI-augmented threats.
Is this trend likely to accelerate?
Given the rapid adoption of AI tools by malicious actors, it is likely that AI-enabled cyber threats will continue to grow more sophisticated and widespread in the near future.
Source: ThorstenMeyerAI.com
