How to Build an AI-First Organization: Lessons from Typeform’s CPO, Aleks Bass

When we talk about the next evolution of product management, “AI-first” has become the buzzword of choice. But as Aleks Bass, Chief Product Officer (and interim Chief Product & Technology Officer) at Typeform, puts it, becoming an AI-first organization isn’t about sprinkling AI features into your product. It’s about redesigning how your organization thinks, operates, and learns. 

In a recent episode of Productside Stories, Aleks broke down what it really means to build an AI-first culture from the inside out — one where every team uses AI intentionally to move faster, make smarter decisions, and stay aligned with real customer outcomes. 

Here’s what product leaders can take from her playbook. 

Redefining What It Means to Be “AI-First” 

Before diving into strategy, Aleks and her team at Typeform spent time defining what AI-first even meant for them. Their conclusion: there’s a crucial difference between being AI-native and being AI-first. 

AI-native refers to products that embed AI directly into the user experience, like an assistive copilot that feels natural and invisible. 

AI-first, on the other hand, is about how your organization works: every workflow, every process, every team using AI to improve decision speed and quality. 

For Typeform, this definition created shared language and focus. “The biggest failure mode,” Aleks said, “is when teams are working off different definitions. You end up with fragmented investments and misaligned expectations.” 

So, step one for any product leader trying to build an AI-first organization is clarity. Define it. Write it down. Socialize it. 

The Human Foundation Behind AI Transformation 

If there’s one misconception Aleks wants to challenge, it’s that AI transformation is a technology problem. It’s not. It’s a people and trust problem. 

Before Typeform could scale its AI ambitions, the leadership team invested heavily in psychological safety and cross-functional trust across product, design, and engineering. Why? Because when teams don’t trust each other, they stay in each other’s lanes — policing decisions instead of collaborating on outcomes. 

That’s where context engineering becomes a hidden superpower for modern product teams. It’s not just about teaching AI models to understand context; it’s about engineering shared context between humans. When PMs, designers, and engineers understand each other’s constraints and trade-offs, decision-making accelerates. 

Aleks put it plainly: “Peak performance comes when those three functions can negotiate the right outcome for the customer, not just their craft.” 

For PMs building AI-first organizations, start here: engineer context and trust before you engineer code. 

Procurement Is Your New Best Friend 

One of the most surprising insights from Aleks’s journey is that your most important partnership in an AI-first organization might not be in product or engineering. It’s in procurement. 

As AI tools exploded, Typeform found itself testing 10–20 solutions at a time, each with different models, pricing structures, and privacy implications. The standard procurement process simply couldn’t keep up. 

So, they created a parallel track: a lightweight, safe “POC (Proof of Concept) procurement” process that let teams experiment quickly while maintaining governance. They limited access, avoided sensitive data, and measured each tool against standardized criteria like model quality, latency, and cost efficiency. 

For Aleks, this was critical: “If you don’t rethink procurement, you’ll spend months evaluating tools instead of learning from them.” 

The takeaway: building an AI-first organization means re-architecting not just your tech stack but your decision stack  (and that includes finance, legal, and compliance). 

Clarity + Observability = Control 

In AI work, chaos loves a vacuum. Without clear strategy and observable metrics, experimentation devolves into noise. 

Typeform tackled this by establishing two principles for every AI initiative: clarity and observability. 

• Clarity: Every team must be able to articulate their AI strategy in a sentence. What problem are we solving? Is it about cost savings, quality, or speed? 
• Observability: Every initiative must be measurable. That means tracking usage, quality improvements, and business impact. 

This level of structure prevents the common failure mode Aleks calls “AI adoption as chaos”: scattered pilots that never scale. 

If you’re leading a product org, bake these two principles into your operating rhythm. Without clarity and observability, your AI-first organization will stall before it scales. 

From Experiments to Outcomes 

One of the strongest signals of AI maturity at Typeform has been speed. Not theoretical speed, but measurable acceleration in real work. 

After enabling engineers with AI coding assistants, the number of PRs submitted per engineer increased by 200%. Integrations that once took a quarter now take a few hours. The design team even connected their design system to an MCP (multi-component platform) that automatically writes 80% of the necessary code for a new feature. 

That’s not about automating people out of the equation. More importantly, it’s about boosting their impact. 

Aleks calls this the invisible value of AI: when the tech disappears and outcomes accelerate. In an AI-first organization, AI becomes a quiet multiplier instead of a headline feature. 

The PM’s Playbook for Becoming AI-First 

So, what can PMs take from Typeform’s experience? Aleks left us with a blueprint that’s as practical as it is powerful: 

1. Define the terms. Be explicit about what “AI-first” means for your company and make sure every team can repeat it back. 
2. Invest in observability early. You can’t improve what you can’t measure. Define how you’ll track AI’s impact before you start. 
3. Lead with trust. Give teams permission to experiment within clear guardrails. The best innovation happens inside safe boundaries. 
4. Treat procurement as a design partner. Rethink how your organization tests and approves tools. It’s now part of product ops. 
5. Engineer context, not control. Your role isn’t to dictate AI decisions but to connect the dots between people, data, and outcomes. 
6. Start small, scale intentionally. Run limited POCs, measure ROI, and expand only what works. 

The Future of AI-First Product Management 

At its core, the shift to an AI-first organization is in amplifying human intuition. AI gives PMs superpowers, but those powers mean nothing without strategy, clarity, and empathy. 

As Aleks said, “AI transformation isn’t about flashy demos. It’s about making AI that invisible accelerant, that 10× investment in everything your company does.” 

For product leaders, the question isn’t if you’ll build an AI-first organization. It’s how fast you can learn to lead one. 

Key Takeaways: 

• Becoming an AI-first organization starts with defining what “AI-first” means for your teams. 
• Trust and shared context between PMs, design, and engineering are the foundation for speed. 
• Procurement and governance are strategic enablers, not blockers. 
• Observability turns AI chaos into measurable progress. 
• The best AI for product managers is the kind that disappears into better outcomes. 

Turn Vision into an AI-First Reality 

• Listen to the full conversation with Aleks Bass and Rina Alexin on Productside Stories — available now on Spotify, Apple Podcasts, and Amazon Music. You can also watch it on YouTube for the complete interview. 
• Take your understanding of AI-first organizations even further with our AI Product Management Certification — built to help product managers turn AI theory into measurable business impact. 
• Now it’s your turn. How is your company evolving toward an AI-first mindset? What experiments, tools, or frameworks are helping your teams bridge strategy, data, and intelligence?
  Share your thoughts and tag @Productside on LinkedIn. We’d love to see how you’re leading the next wave of AI-driven product transformation.