AI Threat Assessment · 26 May 2026

Skydda

AI Cybersecurity
COOKED
6.7/ 10

“Skydda built an AI-powered cybersecurity platform that promises to detect threats faster than human analysts — which is genuinely impressive until you realize they're selling AI to solve problems that other AI is creating at exponentially faster rates. It's like opening an umbrella factory during a hurricane: technically useful, but the weather system you're protecting against scales faster than your manufacturing capacity.”

Business Model
7.5
Automation Risk
6.0
Moat Strength
5.5
Adaptability
7.0
Need Survival
7.5
AI Threat Level
BUSINESS MODEL REPLACEABILITY

Their pitch deck probably says 'we use AI to stop cyber attacks' — which sounds cutting-edge until you notice that Claude, GPT-4, and every security vendor's 'AI-enhanced' product now does threat detection as a feature, not a platform. The cybersecurity-as-a-service model works until the service becomes a commodity API call.

7.5
WORKFORCE AUTOMATION RISK

Security analysts and incident response specialists are being replaced by AI agents that never sleep, never miss a log entry, and don't need equity packages — but someone still has to configure, tune, and explain to the CISO why the AI flagged the CEO's LinkedIn activity as suspicious.

6.0
MOAT STRENGTH

Proprietary threat intelligence and custom detection models could be a real moat — if they weren't competing against OpenAI's next model trained on every CVE database and security blog post ever published. The moat is only as deep as the data advantage, and that advantage shrinks with every foundation model update.

5.5
AI ADAPTABILITY SIGNALS

Building an AI company to fight AI-enabled threats is strategically sound in theory; in practice, it means racing against adversaries who upgrade their capabilities by downloading the same foundation models you're using, except they don't need enterprise sales cycles or compliance frameworks.

7.0
WILL THE NEED SURVIVE AI?

Cybersecurity will always exist, but 'AI-powered threat detection' is rapidly becoming table stakes — like claiming your car has wheels. The question isn't whether companies need security; it's whether they need a specialized AI security vendor or just security vendors with AI built in.

7.5
Verdict

“Skydda faces the peculiar challenge of being an AI company selling protection against AI threats to customers who are simultaneously buying AI tools that make Skydda's AI less special. They're selling premium umbrellas in a market where every weather app now includes rain protection — and the storm they're protecting against keeps learning from their own playbook.”

Scores are based on public information and AI analysis. This is an affectionate roast, not a financial assessment. The best companies use this as a mirror, not a verdict.

Roast another →