CloudSec Academy

Learn about the most pressing security risks shared by all AI applications and how to mitigate them.

Data exfiltration is when sensitive data is accessed without authorization or stolen. Just like any data breach, it can lead to financial loss, reputational damage, and business disruptions.

Greg Zemlin

March 18, 2025

Privilege escalation is when an attacker exploits weaknesses in your environment or infrastructure to gain higher access and control within a system or network.

Lateral movement is a cyberattack technique used by threat actors to navigate a network or environment in search of more valuable information after gaining initial access.

A brute force attack is a cybersecurity threat where a hacker attempts to access a system by systematically testing different passwords until a correct set of credentials is identified.

An attack surface is refers to all the potential entry points an attacker could exploit to gain unauthorized access to a system, network, or data.

Greg Zemlin

March 15, 2025

Cryptojacking is when an attacker hijacks your processing power to mine cryptocurrency for their own benefit.

Remote code execution refers to a security vulnerability through which malicious actors can remotely run code on your systems or servers.

Malicious code is any software or programming script that exploits software or network vulnerabilities and compromises data integrity.

Shadow AI is the unauthorized use or implementation of AI that is not controlled by, or visible to, an organization’s IT department.

A security misconfiguration is when incorrect security settings are applied to devices, applications, or data in your infrastructure.

Shadow IT is an employee’s unauthorized use of IT services, applications, and resources that aren’t controlled by—or visible to—an organization’s IT department.

Uncover the top cloud security issues affecting organizations today. Learn how to address cloud security risks, threats, and challenges to protect your cloud environment.

Shaked Rotlevi

October 25, 2024

Cross-site request forgery (CSRF), also known as XSRF or session riding, is an attack approach where threat actors trick trusted users of an application into performing unintended actions.

Shaked Rotlevi

October 25, 2024

Data sprawl refers to the dramatic proliferation of enterprise data across IT environments, which can lead to management challenges and security risks.

In this blog post, we’ll explore security measures and continuous monitoring strategies to prevent these leaks, mitigating the risks posed by security vulnerabilities, human error, and attacks.

LLM models, like GPT and other foundation models, come with significant risks if not properly secured. From prompt injection attacks to training data poisoning, the potential vulnerabilities are manifold and far-reaching.

Shaked Rotlevi

September 19, 2024

Data leakage is the unchecked exfiltration of organizational data to a third party. It occurs through various means such as misconfigured databases, poorly protected network servers, phishing attacks, or even careless data handling.

Shaked Rotlevi

September 13, 2024

ChatGPT security is the process of protecting an organization from the compliance, brand image, customer experience, and general safety risks that ChatGPT introduces into applications.

Adversarial artificial intelligence (AI), or adversarial machine learning (ML), is a type of cyberattack where threat actors corrupt AI systems to manipulate their outputs and functionality.

Shaked Rotlevi

August 5, 2024

LLM jacking is an attack technique that cybercriminals use to manipulate and exploit an enterprise’s cloud-based LLMs (large language models).

Credential access is a cyberattack technique where threat actors access and hijack legitimate user credentials to gain entry into an enterprise's IT environments.

Shaked Rotlevi

August 2, 2024

Prompt injection attacks are an AI security threat where an attacker manipulates the input prompt in natural language processing (NLP) systems to influence the system’s output.

Data poisoning is a kind of cyberattack that targets the training data used to build artificial intelligence (AI) and machine learning (ML) models.

Dark AI involves the malicious use of artificial intelligence (AI) technologies to facilitate cyberattacks and data breaches. Dark AI includes both accidental and strategic weaponization of AI tools.