![Davos Innovation Week 2026: Crypto and Security Without Compromise]({"desktop":"assets\/images\/cache\/assets\/images\/discovery\/48-1-banner-554x277.png","desktop2x":"assets\/images\/cache\/assets\/images\/discovery\/48-1-banner-1106x553.png","tablet":"assets\/images\/cache\/assets\/images\/discovery\/48-1-banner-380x190.png","tablet2x":"assets\/images\/cache\/assets\/images\/discovery\/48-1-banner-758x379.png","mobile":"assets\/images\/cache\/assets\/images\/discovery\/48-1-banner-560x280.png","mobile2x":"assets\/images\/cache\/assets\/images\/discovery\/48-1-banner-1118x559.png"})
The Datami team took part in Davos Innovation Week 2026, which took place on January 19–23, and presented its expertise. Datami CGO Oleksii Lavrenchuk delivered a talk on the transformation of cyber risks.
![Modern LLM Pentesting Goes Far Beyond Prompt Injection]({"desktop":"assets\/images\/cache\/assets\/images\/blog\/baner-20-554x277.png","desktop2x":"assets\/images\/cache\/assets\/images\/blog\/baner-20-1106x553.png","tablet":"assets\/images\/cache\/assets\/images\/blog\/baner-20-380x190.png","tablet2x":"assets\/images\/cache\/assets\/images\/blog\/baner-20-758x379.png","mobile":"assets\/images\/cache\/assets\/images\/blog\/baner-20-560x280.png","mobile2x":"assets\/images\/cache\/assets\/images\/blog\/baner-20-1118x559.png"})
LLM pentesting goes beyond prompt injection: it audits data leaks, permission bypasses, and harmful actions. Discover what a full AI audit must cover.
![What Does AI Penetration Testing Include? A Practical Breakdown of 12 Critical Security Checks]({"desktop":"assets\/images\/cache\/assets\/images\/blog\/baner-554x277.png","desktop2x":"assets\/images\/cache\/assets\/images\/blog\/baner-1106x553.png","tablet":"assets\/images\/cache\/assets\/images\/blog\/baner-380x190.png","tablet2x":"assets\/images\/cache\/assets\/images\/blog\/baner-758x379.png","mobile":"assets\/images\/cache\/assets\/images\/blog\/baner-560x280.png","mobile2x":"assets\/images\/cache\/assets\/images\/blog\/baner-1118x559.png"})
AI Penetration Testing checks not only for prompt injection but also for RAG, memory, agent privileges, and integrations. Find out which 12 tests help identify real risks in AI systems.
![Cybersecurity Risk Self-Assessment: 6 Gaps Fintech Companies Notice Last]({"desktop":"assets\/images\/cache\/assets\/images\/blog\/chatgpt-image-24-lip.-2026-r.-16_41_00-554x277.png","desktop2x":"assets\/images\/cache\/assets\/images\/blog\/chatgpt-image-24-lip.-2026-r.-16_41_00-1106x553.png","tablet":"assets\/images\/cache\/assets\/images\/blog\/chatgpt-image-24-lip.-2026-r.-16_41_00-380x190.png","tablet2x":"assets\/images\/cache\/assets\/images\/blog\/chatgpt-image-24-lip.-2026-r.-16_41_00-758x379.png","mobile":"assets\/images\/cache\/assets\/images\/blog\/chatgpt-image-24-lip.-2026-r.-16_41_00-560x280.png","mobile2x":"assets\/images\/cache\/assets\/images\/blog\/chatgpt-image-24-lip.-2026-r.-16_41_00-1118x559.png"})
Most companies learn about their weak spots not from their own team, but from a due diligence partner or an attacker. We explain which gaps in fintech remain unnoticed the longest and why.
![Why AI Tokens Introduce a New Class of Smart Contract Risks]({"desktop":"assets\/images\/cache\/assets\/images\/blog\/08-554x277.png","desktop2x":"assets\/images\/cache\/assets\/images\/blog\/08-1106x553.png","tablet":"assets\/images\/cache\/assets\/images\/blog\/08-380x190.png","tablet2x":"assets\/images\/cache\/assets\/images\/blog\/08-758x379.png","mobile":"assets\/images\/cache\/assets\/images\/blog\/08-560x280.png","mobile2x":"assets\/images\/cache\/assets\/images\/blog\/08-1118x559.png"})
A smart contract can pass a Solidity audit without a remark and still be exploitable - if part of the decision-making sits with a model. What a real audit with 40 findings reveals, and what to check before listing.
![What is a Cybersecurity Incident?]({"desktop":"assets\/images\/cache\/assets\/images\/blog\/blog_art.n36-banner-v2-554x277.png","desktop2x":"assets\/images\/cache\/assets\/images\/blog\/blog_art.n36-banner-v2-1106x553.png","tablet":"assets\/images\/cache\/assets\/images\/blog\/blog_art.n36-banner-v2-380x190.png","tablet2x":"assets\/images\/cache\/assets\/images\/blog\/blog_art.n36-banner-v2-758x379.png","mobile":"assets\/images\/cache\/assets\/images\/blog\/blog_art.n36-banner-v2-560x280.png","mobile2x":"assets\/images\/cache\/assets\/images\/blog\/blog_art.n36-banner-v2-1118x559.png"})
Cyber incidents have long ceased to be a headache only for large corporations and government institutions. Today, they are a common part of the digital reality faced by companies of all sizes.
![When AI Meets Smart Contracts: How Prompt Injection Creates a New Web3 Attack Surface]({"desktop":"assets\/images\/cache\/assets\/images\/blog\/1en-554x277.png","desktop2x":"assets\/images\/cache\/assets\/images\/blog\/1en-1106x553.png","tablet":"assets\/images\/cache\/assets\/images\/blog\/1en-380x190.png","tablet2x":"assets\/images\/cache\/assets\/images\/blog\/1en-758x379.png","mobile":"assets\/images\/cache\/assets\/images\/blog\/1en-560x280.png","mobile2x":"assets\/images\/cache\/assets\/images\/blog\/1en-1118x559.png"})
A smart contract can pass the most rigorous audit and the product around it can still be exposed. All it takes is an AI reading on-chain text as a command. A new class of Web3 risk, from a real Datami finding.
![Types of Cybersecurity Vulnerabilities: The Most Common and Critical from Datami’s Practice]({"desktop":"assets\/images\/cache\/assets\/images\/blog\/33-1-1-1-554x277.png","desktop2x":"assets\/images\/cache\/assets\/images\/blog\/33-1-1-1-1106x553.png","tablet":"assets\/images\/cache\/assets\/images\/blog\/33-1-1-1-380x190.png","tablet2x":"assets\/images\/cache\/assets\/images\/blog\/33-1-1-1-758x379.png","mobile":"assets\/images\/cache\/assets\/images\/blog\/33-1-1-1-560x280.png","mobile2x":"assets\/images\/cache\/assets\/images\/blog\/33-1-1-1-1118x559.png"})
In this article, we outline the main types of vulnerabilities. Based on the results of our projects, we have also compiled top lists of the most common and the most critical ones.
