Gartner predicts that generative AI (GenAI) will become a critical workforce partner for 90% of companies by next year. In application development specifically, we see developers turning to code assistants like Github Copilot and Google Gemini Code Assist to help them build software at an unprecedented speed. But while GenAI can power new levels of productivity and speed, it also introduces new threats and challenges for application security teams.

Last week, Google launched Project Mariner, an AI Agent built for the browser, based on the updated Gemini 2.0. Project Mariner is built on top of Google Gemini and can be used to browse the web on your behalf, taking commands from within a Chrome extension and performing autonomous tasks on your behalf.

Launched as an internal project by Spotify in 2016, Backstage was released under the Apache 2.0 open source license in 2020 to help other growing engineering teams deal with similar challenges. Backstage aims to provide a consistent developer experience and centralize tools, documentation, and services within a single platform.

In the rapidly changing field of software development, application programming interfaces (APIs) are very powerful tools. They allow different applications to communicate, share data, and collaborate seamlessly, constituting approximately 71% of all web traffic. However, as APIs become more essential to our applications, they also attract cyber threats. In fact, 57% of organizations reported experiencing at least one API-related data breach in the past two years.

Every single day, billions of API calls happen across the internet. Behind your favorite applications, APIs work quietly to move data and connect systems. But with the growing use of APIs, API attacks didn’t just increase – they exploded. Take the Optus breach in September 2022, in which attackers exploited an unprotected API endpoint and accessed the personal data of up to 9.8 million customers, leading to a $10 million fine.

LLMs are based on neural network architectures, with transformers being the dominant framework. Introduced in 2017, transformers use mechanisms called attention mechanisms to understand the relationships between words or tokens in text, making them highly effective at understanding and generating coherent language. Practical Example: GPT (Generative Pre-trained Transformer) models like GPT-4 are structured with billions of parameters that determine how the model processes and generates language.