AI safety researcher Dr. Roman Yampolskiy believes humanity is moving faster toward artificial superintelligence than most people realize—and that we may be dramatically underestimating the risks.

In this episode of Decoded Genius, Roman explains why many of the world’s leading AI researchers are concerned about the future, whether superintelligent AI can ever truly be controlled, and what happens if we create something smarter than ourselves.

ABOUT THE GUEST
Dr. Roman Yampolskiy is a tenured computer scientist at the University of Louisville, founder of the Cyber Security Lab, and one of the earliest researchers in the field of AI Safety. He has authored more than 100 academic publications and several books on artificial intelligence, cybersecurity, and existential risk.

WHAT WE DECODED
→ Why some AI experts believe superintelligence may be impossible to control
→ The real difference between AI helping humanity and replacing it
→ Whether AI consciousness is possible—and why it matters
→ The timeline for AGI and what could happen next
→ Why the biggest risk may not be malicious AI, but highly competent AI
→ The questions society should be asking before it’s too late

CHAPTERS
00:00:00 Introduction
00:00:39 Why Superintelligence Could Mean Human Extinction
00:04:26 The Oppenheimer Moment and the Money Behind AI
00:07:22 Manipulation, Control, and the Superintelligence Playbook
00:09:19 The 99.9% Risk and Why Control May Be Impossible
00:12:23 Narrow vs General AI: What Builders Could Do Differently
00:13:43 A Weapon of Mass Destruction and the Global Arms Race
00:16:09 Job Loss, Universal Basic Income, and Safe Careers
00:19:39 The AI Tools Paradox and Big Tech's Real Economics
00:22:43 Post-Labor Economics and the Race to AGI
00:26:20 The Next Five Years and Whether AI Can Be Conscious
00:30:33 The Personal Toll of Studying the Last Invention
00:35:22 Is College Still Worth It?
00:38:28 Markets, Bitcoin, and the Agent Economy
00:41:01 Are We Living in a Simulation?
00:46:36 Who's to Blame and the Final Warning

CONNECT WITH ROMAN
Website: https://www.romanyampolskiy.com/
X: https://x.com/romanyam
LinkedIn: https://www.linkedin.com/in/romanyam

FOLLOW DECODED GENIUS
Website: https://decodedgenius.com/
Email updates: https://decodedgenius.com/subscribe/

FOLLOW WILL CANNON
Instagram: https://www.instagram.com/iamwillcannon/
LinkedIn: https://www.linkedin.com/in/will-cannon/
Website: https://iamwillcannon.com/

BECOME A SPONSOR:
Email: will@decodedgenius.com 

#DecodedGenius #ArtificialIntelligence #AI #AGI #RomanYampolskiy

AI safety researcher Dr. Roman Yampolskiy believes we're racing toward superintelligence faster than we realize—and badly underestimating the risk. In this trailer for his upcoming Decoded Genius episode, get a preview of why some of the world's leading AI researchers fear what's coming, whether superintelligence can ever truly be controlled, and what happens if we build something smarter than ourselves.

Full episode dropping soon.

ABOUT THE GUEST
Dr. Roman Yampolskiy is a tenured computer scientist at the University of Louisville, founder of the Cyber Security Lab, and one of the earliest researchers in the field of AI Safety. He has authored more than 100 academic publications and several books on artificial intelligence, cybersecurity, and existential risk.

CONNECT WITH ROMAN
Website: https://www.romanyampolskiy.com/
X: https://x.com/romanyam
LinkedIn: https://www.linkedin.com/in/romanyam

FOLLOW DECODED GENIUS
Website: https://decodedgenius.com/
Email updates: https://decodedgenius.com/subscribe/

FOLLOW WILL CANNON
Instagram: https://www.instagram.com/iamwillcannon/
LinkedIn: https://www.linkedin.com/in/will-cannon/
Website: https://iamwillcannon.com/

BECOME A SPONSOR:
Email: will@decodedgenius.com

#DecodedGenius #ArtificialIntelligence #AI #AGI #RomanYampolskiy