Intro The fear of an evil Artificial Intelligence capable of world domination is a fascinating. It has been the theme of many works of fiction, but as AI becomes more complex and increasingly difficult for humans to fully understand them, a…
Abuse: Prompt Injection
Generative AI is everywhere. It has become so widespread that it is present on virtually every industry and its adoption is still growing and we are already seeing LLMs being used in mass. Software companies are introducing them as a production multipliers. Doctors use them to summarize information about patients allowing them treat even more patients.
Poisoning: Data Poisoning
Nowadays, most AI and Machine Learning algorithms leverage large amounts of data which can be purchased, collected or sourced online. This data is known as the Training Dataset and it enables the model to learn patterns and relationships within the data. By doing so, the model can make predictions or decisions based on new and unseen data.
Poisoning: Supply Chain Attacks
Researchers uncovered more than 100 malicious AI Models on the popular and open-source AI platform, Hugging Face. Many of these were made possible using Supply Chain Attacks. But what is thesupply chain? It’s all of the external parts that an organization relies on to operate.
Privacy: Membership Inference Attacks
AI models are continually advancing their ability to detect patterns becoming increasingly more complex. As AIs are trained on more and more sensitive datasets, it becomes essential to ensure that these models are privacy oriented.