ON DEMAND

Beyond the Safeguards: Exploring the Security Risks of ChatGPT and Other Generative AI Applications​

14 Sep 2023

Gadi Bashvitz

Co-founder & CEO

Gadi Bashvitz

Co-founder & CEO

Gadi Bashvitz is the CEO of Bright Security. He is a highly effective executive with 20+ years of experience holding product, marketing, customer success, and sales positions. With experience selling and partnering with enterprises globally; he has founded, successfully grown, and sold multiple companies in various industries. He started his career in Cyber more than 25 years ago in the 8200 unit of the IDF and continued in various Cybersecurity roles ever since. Graduated from Tel Aviv University with a BA in Information Systems & Management & from the NYU Stern School of Business with an MBA, Magna Cum Laude.

Bar Hofesh

Co-founder & CTO

Bar Hofesh

Co-founder & CTO

Co-founder of Bright Security, Bar serves as their CTO. A globally recognized security and technology expert, Bar has held numerous roles, including CISO, System Architect, Security Advisor, and DevSecOps Advisor at over ten companies. As a leader and researcher, he has contributed to multiple publications and projects in cybersecurity. Bar holds CISO and MCITP certifications.

Daryan Dehghanpisheh

Co-founder

Daryan Dehghanpisheh

Co-founder

Daryan is a senior executive with 20+ years of experience in technology & strategy. He has a demonstrated history of success leading teams in industries ranging from telecom to finance. Highly skilled and qualified in Management, Leadership, Strategy and Technology, he is currently leading and hiring experts in AI and ML, industry leading business development experts, and digital marketing mavens.

Large language models (LLMs) like OpenAI’s ChatGPT have transformed communication and productivity, holding significant potential. However, even with safeguards, LLMs face attacks trying to bypass defenses.

In this webinar we will explore the OWASP Foundation’s Top 10 for LLMs, focusing on prompt injection attacks. We’ll demonstrate safeguard effectiveness with ChatGPT and reveal how prompt injections can bypass protections using various techniques. By shedding light on these vulnerabilities and discussing countermeasures, we aim to deepen understanding of LLM safety and Security Challenges.