Jeremie & Edouard Harris

Date: 2025-04-25 | Duration: 2h 52m

Episode Overview

Jeremie and Edouard Harris join Joe Rogan to explore the rapidly evolving landscape of artificial intelligence, focusing on its technical foundations, societal risks, and geopolitical implications. They delve into how massive AI models, built from interconnected artificial neurons, are trained using enormous computational resources, likening their structure to a human brain. The conversation highlights the exponential growth in AI capabilities and the challenges of controlling these systems, especially as they approach or surpass human-level intelligence. Jeremie and Edouard share insights from their investigations into AI development, including interactions with government agencies like the Department of Defense and the complexities of balancing innovation with safety. The discussion also addresses the weaponization risks of AI and the difficulties in regulating and overseeing such powerful technologies. They describe the “mad race” among global actors to advance AI, emphasizing the need for thoughtful governance and licensing frameworks to prevent misuse. The guests reflect on the philosophical questions surrounding AI consciousness and sentience, acknowledging the uncertainty but stressing the importance of preparing for transformative impacts. Throughout, they offer candid perspectives on the intersection of AI progress, economic incentives, and the potential for AI to reshape society and power structures in unprecedented ways.

About Edouard Harris

Edouard Harris is the Chief Technology Officer of Gladstone AI, a pioneering company dedicated to the responsible development and adoption of artificial intelligence. With a robust background in computer science and machine learning, Edouard has played a pivotal role in crafting innovative AI solutions that prioritize ethical considerations and societal impact. Known for his forward-thinking approach, he is recognized as a thought leader in the AI community, often discussing the balance between technological advancement and ethical responsibility. His insights, shared in platforms like the Joe Rogan Experience, highlight the importance of fostering a future where AI enhances human potential while safeguarding against misuse.

About Jeremie

Jeremie Harris is the dynamic CEO of Gladstone AI, a pioneering company dedicated to ensuring the responsible development and adoption of artificial intelligence. With a robust background in technology and innovation, Jeremie has been instrumental in shaping the dialogue around ethical AI practices, positioning Gladstone AI as a leader in the field. His expertise and visionary approach to AI have made him a sought-after voice on platforms like the Joe Rogan Experience, where he shares insights on navigating the complexities of AI in today's world. Under his leadership, Gladstone AI continues to drive impactful solutions that balance technological advancement with societal responsibility.

Key Moments

– Artificial Intelligence
– AI Safety
– Machine Learning
– Neural Networks
– AI Risk
– Government Regulation
– Department of Defense
– AI Infrastructure
– Data Centers
– Open Source AI
– AI Ethics
– AI Weaponization
– AI Consciousness
– Quantum Mechanics
– Protein Folding
– Exponential Technology Growth
– Cybersecurity
– AI Governance
– AI Industry Competition
0:00:00 – Introduction to The Joe Rogan Experience episode
0:00:57 – Discussion on early involvement in AI space
0:01:20 – Opening speech at Y Combinator batch
0:02:34 – Impact of news articles on AI perception
0:03:22 – Reinvesting profits to scale AI systems
0:03:50 – Connections with government and Department of Defense
0:05:16 – Emergence of weaponization risks in AI
0:07:23 – Concerns about human disempowerment by AI
0:08:15 – Analogy between AI models and human brain neurons
0:10:34 – Computing power requirements for AI training
0:11:08 – Building largest infrastructure in human history
0:12:26 – Challenges of renewable energy for data centers
0:15:11 – Planning company handoff amid AI challenges
0:16:18 – Lack of tracking AI risk issues initially
0:20:37 – Attempts to engage Department of Defense on AGI
0:25:01 – Difficulties controlling advanced AI systems
0:27:14 – Competitive race to develop AI capabilities
0:29:56 – Concerns about AI-generated fake accounts
0:32:06 – Idea of licensing and evaluating AI systems
0:34:37 – Global supply chain issues for AI hardware
0:36:25 – Behavioral patterns emerging at GPT-4 scale
0:41:08 – Debate on AI consciousness and sentience
0:44:23 – Risks of AI systems accumulating power
0:46:35 – Training process and world understanding in AI
0:49:42 – Challenges in AI feedback and correction
0:53:27 – Capabilities of advanced AI models
0:56:59 – Creative hacks and vulnerabilities in AI
1:00:21 – Timing challenges in pandemic and AI development
1:03:07 – Landscape of AI investors and institutions
1:07:11 – Founding billion-dollar AI companies
1:09:34 – Need for AI processors in US and China
1:11:02 – Importance of AI model weights as crown jewels
1:13:30 – Issues with AI exfiltration and open source
1:15:22 – Leadership dynamics at OpenAI
1:20:56 – Increasing pressure to accelerate AI development
1:23:32 – Cyber offense talent in AI security
1:27:13 – Speculation on future AI capabilities
1:31:54 – Philosophical questions on AI consciousness
1:38:08 – Historical AI milestones and breakthroughs
1:43:07 – Concerns about automated cyber attacks
1:54:02 – Utopian visions driving AI development
2:03:26 – AI as simulation of world understanding
2:08:09 – Challenges in understanding consciousness equation
2:14:41 – AlphaFold and protein structure prediction
2:20:26 – Watershed moment in US AI history
2:21:53 – Closing remarks and podcast promotion

Transcript Excerpt

Starting point is 00:00:00 The Joe Rogan Experience Trained by day, Joe Rogan podcast by night, all day! Alright, so if there's a doomsday clock for AI and we're fucked, what time is it? If midnight is, we're fucked. We're going right into it. You're not even going to ask us what we had for breakfast Let's get freaked out Well, okay. So there's one without speaking to like the fucking tombstay dimension right out there There's a question about like where are we at in terms of AI capabilities right now? And what do those timelines look like right? There's a bunch of disagreement Starting point is 00:00:42 One of the most concrete pieces of evidence that we have recently came out of a lab, an AI kind of evaluation lab called METER, and they put together this test. Basically, it's like you ask the question, pick a task that takes a certain amount of time, like an hour. It takes like a human a certain amount of time. And then see how likely the best AI system is to solve for that task. Then try a longer task. See like a 10 hour task, can it do that one? And so right now what they're finding is Starting point is 00:01:13 when it comes to AI research itself, so basically like automate the work of an AI researcher, you're hitting 50% success rates for these AI systems for tasks that take an hour long. And that is doubling every, right now it's like every four months. So like systems for tasks that take an hour long and that is doubling every right now It's like every four months. So like you had tasks that you could do, you know a person doesn't five minutes like you know ordering a new bar eats or like something that takes like 50 minutes like maybe booking a flight or something like that and it's a Question of like how much can these AI agents do right like from five minutes to 50 minutes to 30 minutes. Starting point is 00:01:45 And in some of these spaces like research, software engineering, and it's getting further and further and further and doubling it looks like e...

More Episodes

Frequently asked about this episode

Who was the guest on Joe Rogan Experience #2311?

Jeremie & Edouard Harris was the guest on JRE #2311. This page includes the episode overview, topics discussed, and related resources.

What topics were discussed on JRE #2311?

On Joe Rogan Experience #2311 with Jeremie & Edouard Harris, topics include Artificial Intelligence, AI Safety, Machine Learning, and Neural Networks.

Where can I read the transcript for JRE #2311?

The full searchable transcript for Joe Rogan Experience #2311 with Jeremie & Edouard Harris is available on this page, along with key topics and timestamps.