At a glance
WHAT IT’S REALLY ABOUT
Zapier CEO grades AI PM fluency: from PRDs to factories
- Zapier’s updated AI fluency rubric grades PMs from unacceptable to transformative based on concrete behaviors, with basic AI drafting now considered table stakes in 2026.
- Wade Foster critiques a PRD-only, ChatGPT-assisted spec as outdated and shows that stronger AI-PM work pairs prototypes with rapid, AI-assisted customer evidence gathering.
- The episode frames AI output quality as a judgment problem—poor judgment creates “AI slop,” while good judgment turns AI into a “turbo brain,” and teams need norms that disclose verification effort.
- Transformative AI fluency is depicted as building a product ‘factory’ of agents, memory, and gates that compresses idea-to-shipping cycles, though the optimal system design is still an open question.
- Zapier’s strategic response to new automation competitors is to become more headless and agent-friendly, emphasizing deterministic execution, governance, and an unmatched integrations library.
IDEAS WORTH REMEMBERING
5 ideasAI fluency is measured by behaviors and leverage—not tool familiarity.
Foster’s V2 rubric ranks people by observable behaviors: (1) basic AI usage (summaries/first drafts), (2) structured, reusable workflows and faster iteration, (3) system-building that changes how the org ships. Zapier updates the rubric frequently because capabilities and expectations move quickly.
High-performing PM orgs live in the adaptive middle, not perpetual ‘transformative.’
Zapier expects most PMs to operate “adaptive” most of the time, with occasional “transformative” spikes. Constantly chasing transformative can become counterproductive because you spend all your time tuning systems instead of delivering outcomes.
In 2026, PRD-only work is insufficient; prototype + evidence is the new baseline.
A PRD that’s mostly AI-generated reads as “old-fashioned” in 2026 unless it’s paired with a prototype and customer evidence. The CEO lens is: how far did you push the idea (prototype), and how grounded is it (signals from calls/tickets/forums/data)?
You can delegate the work to AI, but not the accountability—or the judgment.
Foster distinguishes delegating work from delegating responsibility: AI can draft, search, and build—but the PM must apply taste/judgment, filter slop, and stand behind decisions. Teams are developing new etiquette: explicitly label effort/verification level when sharing AI-assisted work.
Transformative is systems design: building the loop that builds product.
Transformative PM work starts to resemble a “product factory”: agents ingest signals, cluster insights, generate specs, prototype, create evals, draft code/PRs, and feed learnings back into memory—with human gates placed intentionally. The hard, unsolved problem is designing the loop (where gates go, cost vs speed, and reliability).
WORDS WORTH SAVING
5 quotesThere is still something to be said about, like, really understanding the customer problem, really having some intuition around how to go solve those problems. And if you combine those, you get excellence. But if you just let AI kinda just do everything, you're probably gonna get kinda mediocrity. You're gonna get what people call AI slop.
— Wade Foster
The line we use at Zapier on slop, you can delegate the work, but you can't delegate the accountability.
— Wade Foster
You probably want to be, in your discipline, you probably want to be farther ahead than your CEO is.
— Wade Foster
It's almost like you're, you're witnessing a new species being invented.
— Wade Foster
To me, transformative is like you are literally an N of one. You're the first of its kind doing this new thing, uh, and you're figuring it out.
— Wade Foster
High quality AI-generated summary created from speaker-labeled transcript.
