“AGI is not here yet, and it's silly for folks to say it is.”
The speaker argues that AGI has not arrived yet, despite claims from some commentators. They discuss two key observations: large enterprises face significant human-centered change management challenges when adopting AI, while true tech companies are seeing real productivity gains from embracing AI development culture.
Summary
The speaker opens with a clear stance that AGI is not here yet and finds it somewhat foolish for people to claim otherwise. They then break their observations into two parts.
The first part focuses on large, established companies and what the speaker calls the 'autonomous enterprise.' They argue that the biggest challenge for these organizations is not technology itself, but change management — specifically, navigating the complex human dynamics involving middle managers, technocrats, and bureaucrats. The speaker emphasizes that many of the internal processes at these companies are not even documented, making rapid AI-driven transformation theoretically appealing but practically very difficult. The human element is identified as the core bottleneck.
The second part shifts to true tech companies, which the speaker describes as energized and culturally aligned with AI adoption. These companies have pivoted to a pro-AI development culture and are seeing tangible benefits: faster development cycles, improved productivity, and accelerated feature rollout schedules.
The speaker closes by pushing back on hype around AGI, calling out those promoting such narratives as essentially 'selling their book.' Drawing on hands-on experience with AI agents and development, the speaker concludes that while current AI tools offer genuine value, they are simply not intelligent enough yet to be considered AGI.
Key Insights
- The speaker argues that AGI has not arrived yet and considers it 'silly' for people to claim it has, particularly those who are perceived as promoting a narrative to sell books or build personal brands.
- The speaker identifies change management — not technology — as the primary obstacle to AI adoption in large enterprises, pointing specifically to middle managers, technocrats, and bureaucrats as the human complexities that slow transformation.
- The speaker notes that many enterprise processes are not even documented, making the theoretical promise of rapid AI-driven transformation much harder to achieve in practice.
- The speaker observes that true tech companies have pivoted their culture to be 'very pro AI development' and are experiencing significantly faster feature rollout and productivity gains as a result.
- Based on hands-on experience with AI agents and development work, the speaker contends that current AI systems offer real value but are simply 'not that smart yet,' directly contradicting AGI claims.
Topics
Full transcript available for MurmurCast members
Sign Up to Access