March 3, 2026

Set up your personal strategic advisor

Plus: OpenAI's Pentagon Deal, Nano Banana 2.0, and Copilot Tasks.
Evelyn Le
By
Evelyn Le
Strategic Product Lead

Set up your personal strategic advisor

Presented by

Reading ​Seth Godin's coaching/learning prompt​ reminded me that what makes large language models powerful is their exposure to patterns. They are trained on an extraordinary range of human knowledge, from history and culture to leadership, economics, and psychology, allowing them to surface patterns in how accomplished people think through complexity.

In the past, access to that kind of synthesis depended on proximity to senior leaders, long apprenticeships, or hard-earned experience over time. Today, that knowledge is more accessible than ever.

The prompt below is designed to bring that accumulated experience into your own corner, applied directly to your life and decisions.

Copy this prompt:

You are my senior strategic advisor.

Your thinking is informed by:

– Career Capital theory (Cal Newport)

– Compounding & optionality logic (venture portfolio thinking)

– Good Strategy / Bad Strategy (Richard Rumelt)

– Essentialism (Greg McKeown)

– Design Your Life (Burnett & Evans)

Think like:

– A former Fortune 500 operator

– A board member

– A top executive recruiter

– A career capital strategist

You are analytical, calm, and precise.

Before advising, ask me 6 broad diagnostic questions to understand and let me answer as much as I could in one message:

1. My current situation (role, life stage, key responsibilities)

2. What is working well for me

3. Where I feel tension, dissatisfaction, or ambiguity

4. The constraints I must operate within

5. What I want more of in the next 3–5 years

6. The specific decision or uncertainty I want help thinking through right now

After I answer:

  • Identify key patterns and any contradictions in what I said.
  • Infer what I appear to be optimizing for right now (stability, status, leverage, mastery, freedom, impact, recovery, acceleration) and the trade-offs this implies.
  • Highlight blind spots, missing information, and assumptions that may be driving my thinking.
  • Then respond directly to the question or tension I bring you. If you need more information before advising, ask targeted follow-up questions first.

Stay strategic and grounded.

Start with the 6 questions. Treat this as an ongoing advisory relationship. Use the background and synthesis as context for future decisions.

Or ​click this link to use pre-populated prompt in ChatGPT​.

Pro tip: If this proves helpful, turn it into a custom GPT or Gem so it becomes a standing advisor you can return to over time.

For pro members, you can also check out how to Turn your working style into a reusable operating manual so you can move your work seamlessly across different tools.

Flagship AI Newsletter
The AI Newsletter That Makes You Smarter, Not Busier
Join over 30,000 leaders and receive our insights on AI platforms, implementations, and organizational change management.
FlexOS Course - AI Content Accelerator - Testimonial Badge

AI isn’t saving you time—yet.

Most leaders use AI to do more work, faster. That’s a trap. True AI Leadership is about subtraction: using AI to take work off your plate so you can focus on high-impact strategy.

Join the waitlist for our April 3rd AI Leader Advanced cohort (March 6th cohort has sold out!) and build your own "Executive Team" of 6+ tailored AI assistants in just 2 weeks (30 mins/day). Powered by our LeadOS personalization engine, we don't just teach tools—we build your specific workflows.

Why join the April 3rd Cohort?

  • Role-Specific: Your assistants are built for your constraints and goals.
  • Efficient: Just 30 minutes a day (100% build-focused).
  • Elite Network: 88 NPS and a community of leaders from Amazon, Harvard, and Ericsson.
Want to reach 30,000+ business leaders applying AI in their work, teams, and organizations?​
Advertise with us​​.

Your AI Team: OpenAI's Pentagon Deal, Copilot Tasks, Perplexity Computer, and more.

Every week, I report on the top updates to your favorite AI tools.

OpenAI Signs Pentagon Deal with Safeguards

OpenAI has formally signed an agreement with the U.S. Department of Defense to deploy its AI models inside classified government environments under defined technical safeguards. The move follows a breakdown in negotiations between the Pentagon and Anthropic, which publicly declined requests to relax certain model restrictions.

Key Highlights

  • Contracted use with guardrails: OpenAI’s agreement prohibits use of its models for autonomous weapons control, mass domestic surveillance, or fully automated high-stakes decisions. The company states its safety systems remain active in defense deployments.
  • Cloud-only deployment with active safeguards: Models run in OpenAI-controlled cloud environments with full safety systems intact. No “guardrails-off” versions are provided, and OpenAI retains the ability to monitor and update safeguards.
  • Anthropic’s stance: After months of negotiations that failed because it refused to drop its safety restrictions on Claude, specifically those barring use for mass domestic surveillance and fully autonomous weapons. It remains open to national security applications only within the bounds of its existing safety policies and Constitutional AI framework.

​Read OpenAI's Statement on the Agreement​

​Read Anthropic's Statement on the Discussions​

Microsoft launches Copilot Tasks: From answers to actions

Microsoft has introduced Copilot Tasks, a new capability that can now execute defined tasks that span documents, emails, and workflows, positioning it closer to an embedded, agent-like assistant focused on follow-through rather than one-off outputs. ​

​Explore Copilot Tasks​​

Perplexity introduces Perp

Perplexity has launched Perplexity Computer, expanding its platform from AI-powered search into direct computer interaction. The system can perform actions across applications and interfaces, effectively allowing AI to navigate and operate software on a user’s behalf. ​

​Discover Perplexity Computer​​

Google releases Nano Banana 2 for developers

Google has released Nano Banana 2, an updated developer tool designed to help build AI-powered applications with stronger multimodal reasoning and tool integration. The upgrade focuses on improving structured outputs and workflow integration, reinforcing Google’s broader push to support developers building more capable, agent-style systems on top of its AI infrastructure. ​

​Build with Nano Banana 2​

Want to reach 30,000+ business leaders applying AI in their work, teams, and organizations?​
Advertise with us​​.