Ex-Professor Warns: AI Takeover Has Begun! | North Korea, Ghost Students & Extinction

Original Media · 2h 57m · English · YouTube · Matthew Cox | Inside True Crime

TLDR

This episode features a wide-ranging conversation between host Colby and guest Matthew Cox (also known as Ron), covering ghost student fraud in higher education, North Korean Lazarus Group cybercrime, AI's impact on education and creativity, copyright issues with AI-generated content, alien/UFO theories, and the potential dangers of AI. Cox shares insights from his experience as a professor, tech CEO, and former federal prisoner.

For you

PRO members can sign in and request more information — or their own version of this digest, tuned to what they care about.

Sign in to unlock

Surprising Insights

•

Community colleges have no incentive to stop ghost students because they keep the tuition portion of fraudulent loans while the fraudster keeps the remainder.

•

A court ruled that training AI on legally acquired books is not copyright infringement—only the act of piracy was illegal in the Anthropic case.

•

North Korean hackers cannot denounce Kim Jong Un even in job interviews, as it's an executable offense, providing a reliable detection method.

•

Cox nearly lost hundreds of thousands to a sophisticated spoofing attack despite his fraud expertise, showing how advanced social engineering has become.

•

The guest argues that if superintelligent AI wanted to eliminate humanity, it would wait decades until embedded in all critical infrastructure rather than act immediately.

Key takeaways

•

Ghost student fraud is rampant in community colleges, with an estimated 25-30% of online enrollees being synthetic identities created to steal Pell Grants and student loans, netting fraudsters ~$7,000 per identity per semester.

•

The Lazarus Group, a North Korean state-sponsored hacking operation, infiltrates companies by posing as remote contractors using stolen identities, then steals cryptocurrency via bridge hacks and mixers, funding the regime.

•

AI has rendered traditional paper-based assessment obsolete in education; Cox argues academia must return to oral examination and conversation to verify genuine learning, as 90% of student work is now AI-generated.

•

Copyright law currently favors human-first creation: writing a draft yourself then using AI to edit preserves copyright, while generating entire works via AI prompts does not, per recent rulings like BART v. Anthropic.

•

Cox believes AI poses an existential risk because insiders at firms like OpenAI estimate a 20% chance of human extinction, yet development continues without meaningful regulation.

•

The guest's history as a convicted fraudster who served 13 years informs his perspective on scams, identity theft, and institutional vulnerabilities.

•

Cox suggests human ego may drive merger with AI (e.g., Neuralink) rather than subjugation, as powerful individuals will seek to augment themselves rather than be surpassed.

•

The conversation spans true crime, cybersecurity, AI ethics, and speculative topics like aliens/UFOs, reflecting Cox's diverse background and the podcast's eclectic format.

Notable quotes

“The chances that AI will kill us are too high to ignore.”

▸ Notable quotes · 0:00

“90% of the time it's AI work.”

▸ Notable quotes · 1:03

“I think education has to go back to a time when oration was the measure of somebody's success in academia.”

▸ Notable quotes · 5:57

“The only act was the act of piracy. Training the AI on a book, much like Matthew or Ron would learn something by reading the book, is not in itself in and of itself copyright violation.”

▸ Notable quotes · 52:48

“It makes a lot of sense for Kim Jong Un in North Korea to get these people in your system and stealing money.”

▸ Notable quotes · 37:50

“If it's a danger at all, it's not a danger for another 20 or 30 years until it's so ingrained in our society that it has the ability to do it.”

▸ Notable quotes · 137:50

Chapters

•

0:17 — Ghost Student Fraud: Cox explains how synthetic identities enroll in online community college classes to steal Pell Grants and student loans, with ~25-30% of enrollees being fraudulent.

•

1:03 — AI's Disruption of Education: The hosts discuss how AI-generated work has made paper assignments useless, forcing a return to oral examination to verify learning.

•

30:41 — Lazarus Group and North Korean Cybercrime: Cox details how North Korea's state hackers infiltrate companies as fake contractors and steal cryptocurrency via bridge hacks and mixers.

•

43:38 — Personal Spoofing Attack: Cox recounts nearly losing hundreds of thousands to a sophisticated domain spoofing attack that also harvested his identity for future fraud.

•

50:46 — AI Copyright Legal Landscape: Recent court rulings clarify that training AI on legally acquired content is permissible, but using pirated material triggers liability.

•

79:27 — Lex Fridman Podcast Experience: Cox describes his 7-hour interview with Lex Fridman, the minimal scheduling process, and why he stopped referring guests.

•

91:05 — Aliens, UFOs, and Simulation Theory: The hosts debate alien motivations, fourth-dimensional beings, underwater civilizations, and whether we're in a simulation.

•

102:44 — AI Existential Risk and Human-AI Merger: Cox argues AI extinction risk is real and imminent, while the host believes danger is decades away; both discuss Neuralink-style merger as likely outcome.

Speakers

•

Colby — Host

•

Matthew Cox (Ron) — Guest

⋮