Join us for an engaging workshop where we delve into state-of-the-art (SOTA) prompt engineering techniques as featured in "The Prompt Report," a comprehensive 76-page survey paper Sander has done in collaboration with OpenAI and Microsoft.

Additionally, we will explore the security implications of prompt injection discovered during our groundbreaking global competition, HackAPrompt.

Attend this workshop to:

• Gain insights into cutting-edge prompt engineering techniques and their practical applications.
• Understand how these techniques can be tailored to enhance AI model performance across various domains.
• Discuss real-world implications and applications of prompt engineering in different industries.
• Learn best practices for integrating prompt engineering into your AI development workflow.

By participating in this workshop, you will gain a deep understanding of the latest advancements in prompt engineering and how to effectively apply them to drive innovation and performance in AI applications.

In this talk, Sander will discuss the current state of Prompt Engineering drawing from his recent survey paper, The Prompt Report, done in collaboration with OpenAI and Microsoft. The 76 page paper is the largest ever study of prompting, and has gone viral on social media, obtaining millions of views. He will discuss the prompt classification system he developed for prompting techniques, including Zero-Shot, Few-Shot, Decomposition, Thought Generation, Ensembling, and Self-Criticism. Attend this talk to delve into the technical and business implications of prompting and how to get the best results for your teams.