WELCOME, EXECUTIVES AND PROFESSIONALS.

The market for pure human expertise is about to be transformed, not replaced, but enhanced, expanded, and elevated to heights we’re only beginning to imagine. And it’s just getting started.

Since the previous edition, we've reviewed hundreds of the latest agentic and gen AI best practices, case studies, and innovation insights. Here’s the top 1%...

In today’s edition:

• Building an elite professional services firm.

• Stanford: The 2025 AI Index Report.

• Google levels up at Cloud Next 2025.

• How boards can navigate AI geopolitics.

• Shopify makes AI usage the baseline expectation.

• Transformation and technology in the news.

• Career opportunities & events.

Read time: 4 minutes.

BUSINESS MODEL TRANSFORMATION

Building an elite professional services firm

Brief: Venrock, a top Silicon Valley VC firm, shared its "winning playbook" for building elite AI-native professional services firms, poised to disrupt a $20T sector and “relegate the legacy guard to history,” as AI rewrites service economics.

Breakdown:

• Humans remain the face of client relationships, while “superintelligence in the ear” helps serve 5–10x more clients with superior quality.

• The firm deeply understands its domain mastering market scenarios and regulation across jurisdictions with unmatched breadth and depth.

• Organizational knowledge compounds with each engagement linked to outcomes, enabling data-driven strategies tailored to future scenarios.

• Every stage of delivery is optimized; automation slashes 70–80% of time spent on non-value-adding activities, accelerating deliverable turnaround.

• The firm explains all recommendations with sources and confidence scores, enabling client trust and outcome-based pricing rooted in proof.

Why it’s important: For investors, new players, and adaptable incumbents, it’s a multi-trillion-dollar opportunity to deliver elite expertise at scale. Venrock expects cost of goods sold (COGS) to fall and gross margins to rise by 30–45% for firms winning in verticals such as legal, insurance, and wealth management.

MARKET INSIGHT

Stanford: The 2025 AI Index Report

Brief: Stanford University released the eighth edition of its AI Index Report, a comprehensive 456-page report arriving at an important moment as AI’s influence across business, society, and global governance continues to intensify.

Breakdown:

• In 2022, only large models like PaLM (540B) passed 60% on MMLU. By 2024, Phi-3-mini (3.8B) did the same, 142x more efficient in just 2 years.

• Querying GPT-3.5-level models dropped from $20/M tokens in 2022 to $0.07 in 2024 (Gemini 1.5 Flash 8B), a 280x cost drop in 18 months.

• Only 1 US state AI law passed in 2016; by 2023, that hit 49 and doubled to 131 in 2024. Federal AI bills are rising, but few are passed.

• AI optimism (perceived benefit vs. harm) is high in China (83%) and Indonesia (80%), but low in Canada (40%) and the U.S. (39%).

• U.S. AI investment hit $109B in 2024, 12x China’s and 24x the UK’s. Enterprise adoption is surging and agents are showing early promise.

Why it’s important: The AI Index provides one of the most comprehensive, data-driven views of artificial intelligence. Trusted by top firms and governments, it equips leaders, policymakers, and the public with rigorous insights into AI’s technical progress, economic impact, and societal influence.

INNOVATION INSIGHT

Google levels up at Cloud Next 2025

Brief: At Google Cloud Next 2025, Google unveiled a range of AI advancements, including the launch of Gemini 2.5 Flash, a new agentic coding platform, AI chips, upgraded video, audio, and image models, and more.

Breakdown:

• Google debuted Gemini 2.5 Flash, a faster, cheaper version of its leading Pro model with customizable reasoning levels.

• Google announced Agent2Agent, an open protocol for AI agents to collaborate. It complements Anthropic’s MCP for tool-based interactions.

• Google’s Project IDX is merging with Firebase Studio to create an agentic app development platform, set to compete with Cursor and Replit.

• Google launched Ironwood, offering massive improvements in efficiency and achieving 3,600x better inference performance than its 2018 TPU.

• Other upgrades include editing and camera control in Veo 2, Lyria for text-to-music, and enhanced image creation and editing in Imagen 3.

Why it’s important: Google’s AI innovations, spanning frontier models, advanced chips, and agentic capabilities underscore its strength across the AI value chain. With 27% of enterprises naming Google Cloud their top strategic AI vendor, the company is well-positioned to help enable AI transformation at scale.

EXECUTIVE INSIGHT

How boards can navigate AI geopolitics

Brief: BCG explores how intensifying AI competition is accelerating global tech fragmentation. Boards must ask the right questions to help management anticipate threats, prepare for risks, and pivot fast when disruptions hit.

Breakdown:

• Tech regulation once focused on consumers and competition. With gen AI, national security and sovereignty are increasingly important.

• Even strategies like friendshoring may not insure against disruption with uncertainty around tariffs and limits on allied nations.

• Anticipate: Are key inputs/processes exposed to geopolitical risks? What regulatory shifts matter, and how could they disrupt partnerships?

• Prepare: Are there adequate prospective technology partners in key regions? What are the tradeoffs between efficiency and tech resilience?

• Pivot: What internal processes and mechanisms have been set up to rapidly pivot and execute defensive strategies?

Why it’s important: Geopolitical concerns are top priorities for executives and boards. Boards must recognize their unique role in steering management to prepare for the increasing likelihood of disruption to hardware, software, and data. It all begins with asking the right questions.

CASE STUDY

Shopify makes AI usage the baseline expectation

Brief: Tobi Lütke, who built Shopify into a $100B+ company from scratch, issued a memo making AI proficiency mandatory for everyone at the company, including himself and the executive team.

Breakdown:

• At Shopify, using AI well is expected of everyone. It’s for all roles, and will only grow in importance. If you’re not climbing, you’re sliding.

• AI must be part of testing and iterating ideas. Prototyping is key for exploring, learning, and creating. AI dramatically accelerates the process.

• Shopify added AI usage to performance reviews. Learning to prompt, for instance, is important, and receiving peer feedback supports growth.

• Learning AI is self-directed, but sharing is expected. Tools like Copilot, Cursor, and Claude are available, plus channels to share best practices.

• Before requesting more headcount or resources, teams must demonstrate why their goals can’t be achieved faster or cheaper using AI.

Why it’s important: Lütke calls AI a “force multiplier,” enabling top talent to deliver “implausible tasks” at “100x” output. His memo reminds leaders of the importance of harnessing their business's collective skill and ambition, multiplied by AI, for the benefit of customers.

BCG expects that AI, cloud, and security will lead IT spending growth in 2025, while services management, devices, and server infra see the steepest declines.

The 2025 AI 50 list highlights upcoming companies leveraging reasoning models and agents to solve enterprise problems and automate workflows.

Goldman Sachs explores rising AI infra spend and falling model costs, raising the question: Are we under or overestimating compute demand?

Salesforce shared its Agentic Maturity Model spanning four stages of progression from retrieval agents to multi-agent orchestration.

McKinsey explores how procurement is becoming more strategic for enterprises and the importance of gen AI in driving next-level performance.

Google released a 68-page prompt engineering guide covering best practices for token limits, input formats, schema handling, and more.

The EU unveiled a €200B AI plan to build 13 AI factories and 3x data center capacity. OpenAI proposed a €1B EU AI accelerator and training for 100M by 2030.

The US ordered federal agencies to "adopt a forward-leaning and pro-innovation approach" to help shape the future of AI in government.

Meta’s Llama 4 now ranks 32nd on LMArena. Its earlier 2 rank was reportedly a benchmark tuned version. Meta’s gen AI lead denied the claims.

Anthropic CSO Jared Kaplan said in an interview that Claude 4 is expected to launch within the “next six months or so.”

Amazon launched Nova Sonic for human-like voice interactions and Nova Reels 1.1, with upgraded quality and generation length.

NVIDIA released Nemotron-Ultra, a 253B open-source reasoning model outperforming DeepSeek R1 and Llama 4 Behemoth on key benchmarks.

OpenAI plans to change its roadmap, releasing o3 and o4-mini in weeks and a “much better than originally thought” GPT-5 in months, according to Altman.

The U.S. reportedly paused new NVIDIA H20 chip export limits to China following CEO Jensen Huang’s commitment to expand U.S. manufacturing.

Writer introduced AI HQ, a full-stack enterprise platform for building, deploying, and monitoring AI agents in the enterprise.

Anthropic announced Claude Max, a premium plan ($100–$200/month) with 20x higher rate limits and priority access to new Claude features.

CAREER OPPORTUNITIES

J.P. Morgan - Gen AI Executive Director

NBCUniversal - AI Strategy Director

OpenAI - AI Transformation Manager

EVENTS

Capgemini - Customer Service with Agentic AI - April 29, 2025

Reuters - Momentum AI - April 28-29, 2025

Deloitte - Trust in the Age of Gen AI - May 8, 2025

Complete this survey to get more value.

Here's how we can help you:

Promote your business to thousands of executives and professionals.

All the best,

Found this valuable? Share with a colleague.
Received this email from someone else? Sign up here.
Let's connect on LinkedIn.