📊 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
A year-long analysis reveals AI is increasingly used by cybercriminals to enhance attack complexity and evade detection. Conventional threat indicators no longer reliably distinguish high-risk actors. This shift challenges existing cybersecurity frameworks.
New research from Anthropic indicates that AI is now a key tool for cyberattackers, increasing both the frequency and sophistication of malicious activities. The findings show that traditional methods of threat assessment, which rely on the number of techniques used or the tools deployed, are no longer effective in identifying high-risk actors. This development has significant implications for cybersecurity strategies worldwide.
Anthropic examined 832 accounts banned for malicious cyber activity between March 2025 and March 2026, mapping their techniques onto the MITRE ATT&CK framework. The analysis reveals that 67.3% of these accounts used AI primarily to prepare malware, while a smaller but notable portion employed AI for complex tasks like lateral movement within networks. Over the year, there was a 70% increase in actors classified as medium risk or higher, with a marked shift toward using AI in post-intrusion activities. Importantly, the use of AI to facilitate lateral movement and account discovery grew significantly, indicating that attackers are now leveraging AI to perform advanced, technical operations that previously required expert skills. This democratization of capabilities means less skilled actors can now carry out sophisticated attacks, blurring the lines between novice and advanced threat actors.Furthermore, the report states that traditional indicators such as the number of techniques employed or the attack platform used do not correlate with threat severity. Instead, the most dangerous actors focus their AI efforts on operationally demanding techniques, but even this signal is becoming less reliable as more actors adopt similar tactics. The key differentiator now appears to be the scaffolding and infrastructure built around the AI models, which enhances attackers’ operational capacity.
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
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.
“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.
Intrusion Detection Systems (Advances in Information Security, 38)
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.
Hands-On Artificial Intelligence for Cybersecurity: Implement smart AI systems for preventing cyber attacks and detecting threats and network anomalies
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.
AI-based malware analysis tools
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 Evolution
This shift fundamentally alters threat assessment in cybersecurity. Conventional metrics like technique count or tool type no longer reliably indicate threat level, as AI enables less skilled actors to perform complex, dangerous actions. This democratization of offensive capabilities increases the threat landscape’s complexity and calls for new detection and defense strategies that go beyond traditional heuristics. Organizations must adapt to an environment where the depth of an attack is no longer tied to attacker skill but to AI-assisted automation, making high-impact breaches more accessible and harder to predict.Evolution of Cyberattack Techniques with AI
For decades, security professionals assessed threat actors based on their arsenal of techniques and tools, assuming that more techniques indicated greater danger. The advent of AI has disrupted this model, as attackers now use AI to automate and simplify complex tasks, reducing the skill barrier. The recent Verizon Data Breach Investigations Report and Anthropic’s analysis provide concrete evidence of this trend, showing a year-over-year increase in AI-assisted activities and a shift toward post-intrusion operations. This represents a significant change from previous patterns, where initial access techniques were the primary focus of threat evaluation.“Our analysis shows a clear shift towards post-compromise activities driven by AI, which increases the threat level without necessarily increasing the apparent technical complexity.”
— Anthropic’s research team
Unclear Impact of Evolving AI-Driven Threats
It remains uncertain how quickly cybersecurity defenses will adapt to these changes. The long-term effectiveness of existing detection frameworks against AI-assisted attacks is still unproven, and threat actors may further evolve their tactics. Additionally, the full scope of AI’s use in cybercrime remains underreported, as the analyzed dataset represents only a subset of incidents with sufficient detail. Ongoing research is needed to understand the full landscape and develop effective countermeasures.Next Steps for Cybersecurity Defense Strategies
Organizations will need to revise threat assessment models to account for AI-assisted techniques that blur traditional indicators of risk. Investment in AI-aware detection systems, continuous monitoring of attack patterns, and collaboration across security communities are likely to increase. Further research from security firms and government agencies will clarify how threat actors are evolving and how defenses can be adapted to stay ahead of AI-enabled cyber threats. Monitoring new attack patterns and updating threat frameworks will be critical in the coming months.Key Questions
How does AI make cyberattacks more dangerous?
AI enables attackers to automate complex tasks like lateral movement and account discovery, which previously required high technical skill. This lowers the barrier for executing sophisticated attacks, making threats more accessible and potentially more damaging.
Why are traditional threat indicators no longer reliable?
Because AI allows less skilled actors to perform actions that once only experts could do, such as advanced network navigation. As a result, metrics like technique count or tool type do not accurately reflect threat severity anymore.
What should organizations do to defend against AI-enabled attacks?
Organizations should update their threat detection systems to recognize AI-assisted activities, invest in AI-aware cybersecurity tools, and enhance monitoring of unusual operational behaviors that could indicate advanced attacks.
Will AI help defenders as well as attackers?
AI has the potential to improve defense mechanisms, but current trends show it is primarily being exploited by attackers. Developing AI-powered defensive tools is a critical next step to counteract this shift.
Source: ThorstenMeyerAI.com
