AI Safety Is an Engineering Problem, Not A Research Program & Introducing Business 2030

Perspectives on AI safety and alignment, and what you as a business leader need to know.  I’m attaching a good set of resources for you to read, this is a fascinating issue.

Second, I introduce the idea of Business 2030 – the new, ai-powered operating model for business, which is coupled with HR 2030. As you’ll hear, the new world of HR is going to become essential to your company’s AI business operating model in the coming years. So all the details in HR 2030 are urgent.

Reading and Background

AI Doesn’t Need to Slow Down. The AI Industry Needs to Grow Up. (substack)

HR 2030: How AI Superagents Transform HR From Service Delivery to Dynamic Business Enablement

Galileo Pluto Release: Enterprise HR Intelligence in Any AI Agent

“Superpowered,” our new book that explains the positive future of work

AI Alignment Approaches

OpenAI Discloses Six New Incidents of ‘Concerning’ A.I. Behavior

Google DeepMind launches institute to widen the AGI debate

Claude’s New Constitution

Microsoft AI’s “Code of Conduct”

OpenAI’s Model Behavior Spec

Is AI Conscious? (Sam Harris, Cameron Berg – worth watching)

Chapters

• (00:00:00) - Intro: Europe trip and today's topics
• (00:00:39) - AI safety as an engineering problem, not research
• (00:03:01) - Regulatory capture, money incentives, and business legitimacy
• (00:04:16) - Aligning agents as a corporate buyer
• (00:04:56) - HR 2030 reference blueprint and agent architecture
• (00:06:21) - Business 2030: dynamic enablement and AI transformation
• (00:08:08) - Case study: software vs organizational readiness
• (00:09:10) - Superworkers, embedded AI, and idea constraints
• (00:10:46) - HR becoming business-centric
• (00:12:13) - Client example: beverage company restructuring
• (00:13:19) - Career opportunities and upcoming events
• (00:13:49) - Galileo Jupiter release and closing