About this episode

The episode discusses the risks of prompt injection in AI and the significance of ChatGPT's new lockdown mode.

Send us Fan Mail A single “helpful” AI response can turn into a data leak if the model is tricked into following the wrong instructions. That’s the core risk behind prompt injection, one of the most misunderstood attack vectors in modern AI, and it’s why OpenAI’s new ChatGPT lockdown mode matters more than it sounds at first glance. We walk through what prompt injection is, how it can be embedded in everyday inputs like emails, documents, and even images, and why it works so well against lar...

People in this episode

Host: Jim Kunkle

Topics covered

• AI security
• prompt injection
• data leaks
• ChatGPT
• technology risks

Mentioned in this episode

Organizations: OpenAI

Products: ChatGPT

More episodes of The Digital Revolution with Jim Kunkle

• Using AI as a Thinking Partner, Not Just a Task Tool w/Eddie Irvin · July 20, 2026 · 39 min
• America's Surprising AI Power Shift: What The Verasight Survey Tells Us · July 15, 2026 · 7 min
• AI That Works For Accounting Teams w/Peter McCarroll · July 13, 2026 · 40 min
• The New Social Engineering: Why Scammers Don’t Need Your Password Anymore · July 8, 2026 · 7 min
• The Voice That Wasn't Them: AI Cloning Scams And The New Emotional Cybercrime · July 6, 2026 · 25 min
• The 90‑Second Rule for AI‑Native Thinking · July 1, 2026 · 5 min

Explore listener stats, chart rankings, contacts and more on the The Digital Revolution with Jim Kunkle podcast page.