I’ve been straddling AI for some time now between involvement in projects to develop AI products and services and company efforts to work out their AI strategy to add value to their businesses. The tension between investment/spending on AI and getting visible value is pushing the need for disciplined focus. Boards, Private Equity and Senior Teams will do “something” around AI in 2025. The risk of doing more harm than good is real, and you will have a strategy either implicit or explicit; so better to make it explicit.

Programmers are already facing the reality that their world has changed and is accelerating. This should serve as notice to leadership both around the pace and reality of change AND the need think in a disciplined manner around AI. For example the Cursor team was on Lex Fridman recently(its worth a listen). Aman from Cursor commented that during a conversation about the laws of scaling he realized that he was going through the states of grief, denial, anger, etc; and had to get to acceptance to be able to think about what is possible in terms of progress. Just being a few months ahead in terms of action is an advantage but the Cursor team pace is that in a year, their product will make the current version look obsolete.

I am still seeing companies aware of the importance of AI but either too vague about its use or on the other end, boiling the ocean with claims and aspirations of AI changing everything in 2025. Regardless, these are at least some of the questions that require attention as part of your AI path.

1. Do you have a cross functional AI group to guide your AI effort?
   • For example: Jeff Sprecher, CEO at ICE, announced during his Q1 call this year that he was forming an internal AI center of excellence to look at use cases and trends impacting ICE after it came up at a Futures Industry spring industry conference
2. What is your short list of assumptions and use cases across AI themes over the next 24 months company wide and in each functional area?
   • Reporting and Analytics
   • Agentic AI
   • Personalization, Profiling and Identification
   • Coding assist
   • Prediction
   • Automation
   • Content generation
3. How do your assumptions about the state of AI impact the answers your key business strategy questions: (see Roger Martin’s work)
   • What are your aspirations and concrete business objectives?
   • Where do you play and not play?
   • How do you win against the competition?
   • What capabilities do you leverage or need?
   • What management systems need to be put into place; what I call the company operating system
4. Are you getting at the right, few metrics?
   • Real cost of AI
   • Objective improvement of process
   • Evidence of value added/ efficiency developed
   • Iterate your metrics
5. What aspects of AI do you want to drive top down across a group of companies or functions vs. bottom up surfacing of practical use?
6. How can risk management play a positive role in successful implementation? (like in trading, good risk management is often the key to value enhancement)
   • Do you have AI operating principles or governance of use?
   • Do you have a defensive strategy to spot and mitigate nefarious use of AI as part of your cyber strategy
   • What are the gates for moving from POC to implementation
7. How does your AI strategy prioritize against other strategic moves and what can the organization tolerate and be successful?
8. How are you preparing the organization and your people for AI use through real application of the organizational change stack?
   • Alignment
   • Practical Engagement
   • Active Leadership
   • Individual Accountability