AI PM Ready
Back to the Digest

Nilay Patel: AI Costs Are Too High, Benefits Not Real Enough

Simon WillisonProduct Strategy4 min readImportantOriginal source ↗

Nilay Patel argues that current AI products are not yet good enough to justify their costs, citing high energy consumption, unconvincing benefits, and unproven agentic capabilities. He states, 'The products aren't good enough to sell people on the costs.' This perspective challenges the current AI hype cycle and suggests a need for more tangible, user-centric value propositions. PMs must focus on delivering demonstrable ROI and user experience improvements that outweigh the significant costs. This week, re-evaluate the core value proposition of your AI features against their operational costs and user-perceived benefits.

What happened

Nilay Patel, in a discussion on The Vergecast, asserts that the current generation of AI products fails to deliver sufficient value to justify their high costs. He highlights issues such as the energy consumption of data centers, the prevalence of 'hilarious deep fake slop,' and the current immaturity of agents. Patel's core argument is that "The products aren't good enough to sell people on the costs," suggesting that the industry is overspending on AI without delivering commensurate user benefits. He implies that until AI products demonstrably improve users' lives without requiring significant convincing or incurring excessive costs, the industry will struggle to overcome these challenges.

A scenario you might face

A Series B SaaS company is considering a significant investment in integrating advanced AI capabilities into its core platform. While the engineering team is excited about the technical possibilities, leadership is concerned about the projected increase in cloud compute costs and the unproven user adoption of similar AI features in the market. Patel's critique serves as a critical data point, forcing the PM to pivot from showcasing technical prowess to building a compelling business case that clearly articulates the value proposition and ROI to stakeholders, rather than relying on the general AI trend.

What changes in your practice

In your next stakeholder meeting discussing AI feature development, replace discussions about AI capabilities with a clear articulation of the specific user problem being solved and the quantifiable benefits delivered, directly addressing the cost-to-value ratio. Before your next roadmap review, add a 'cost-benefit validation' step for all proposed AI features, requiring evidence of tangible user value that outweighs the projected operational expenses.

Questions for your team

  1. 01What are the key metrics that demonstrate 'real benefits' for our specific AI features, beyond theoretical improvements?
  2. 02How can we quantitatively measure the 'slop' or negative user experiences in our AI products and set targets to mitigate them?
  3. 03What is our strategy for transparently communicating the costs and benefits of AI features to users and stakeholders?
  4. 04What benchmarks or case studies can we point to that demonstrate AI products successfully justifying their costs in our specific industry?

Practise this

Interviewers ask about stories like this one. Rehearse your answer in a timed mock interview.

Start a mock interview →
All stories
Was this page useful?