AI News Weekly - 100 years from now : The Case for Artificial Stupidity - Mar 23rd 2026
Why I picked this
Victor flags this because it inverts the entire AI capability race. While everyone's optimizing for speed and autonomy, this piece asks: what if we're building the wrong thing? The 'artificial stupidity' frame isn't cute contrarianism — it's a serious design question about intentional friction. The author's exploring what happens when we optimize for human agency preservation instead of task completion velocity. This matters now because we're hardcoding automation assumptions into systems that will compound for decades. The philosophical framing ('100 years from now') gives permission to question premises we're treating as axioms in 2025. Worth reading not for predictions but for the design principles it surfaces: when should AI deliberately slow down, ask dumb questions, or force human decision points? That's the kind of systems thinking that separates builders from feature shippers.
Three lenses
The 'worse on purpose' constraint is actually a product spec — I'd prototype an AI assistant that requires human confirmation on every third action, measure task completion vs. error rate, and see if intentional friction creates better outcomes than full automation. Deployable this quarter.
Philosophically interesting, operationally vague. Show me the pilot where 'artificial stupidity' improved win rates or reduced churn, then we'll talk about rolling it out. Until then, this is a dinner party conversation, not a deployment strategy.
Everyone will nod along to this and then immediately go back to automating everything because 'intentional friction' doesn't show up in velocity metrics. The real test: name one company that's actually shipping AI that's deliberately less capable. You can't, because the incentives don't support it.
Why this matters for operators: Surfaces the design question operators aren't asking: when should AI deliberately not automate? Relevant for teams building internal tools where error cost exceeds speed benefit.
I cover AI×GTM intelligence like this every Wednesday.
Get STEEPWORKS WeeklyMore picks
Humans will keep inventing new reasons why we must stay in the loop with agents
- Human resistance to full AI autonomy is not purely technical—it's psychological and organizational; companies will rationalize keeping humans in decision loops even when agents are capable
- The 'human-in-the-loop' requirement may become a self-perpetuating narrative rather than a genuine necessity, driven by organizational risk aversion and change resistance
- Product leaders at scale (Notion) are observing this pattern, suggesting it's a widespread phenomenon across enterprise AI adoption, not isolated to specific use cases
Comp plans for consumption pricing
- Consumption pricing fundamentally breaks traditional SaaS comp models—requires rethinking sales incentive structures around usage vs. contract value
- Four distinct contract structures exist (pay-as-you-go, uncommitted, committed, hybrid), each requiring different compensation mechanics and sales behaviors
- Enterprise consumption-based deals create tension: customers want flexibility, sales teams need predictability for quota attainment—comp design must bridge this gap
3 revenue motions your AI is only half wired into
- Model parity has arrived: OpenAI/Claude now trade evenly on core tasks, making 'better AI' a non-differentiator—the edge shifts to integration depth into existing revenue motions
- Waste is quantified: teams paying $17K-$37K/month for AI seats that never touch pipeline generation; real cost is opportunity cost of unused capacity, not subscription fees
- Lean teams have a structural advantage: cannot out-buy larger competitors on model access, but can out-embed them by wiring AI 1 revenue motion deep (pipeline → content → deals) with proprietary deal context competitors haven't seen
This analysis was produced using the STEEPWORKS system — the same agents, skills, and knowledge architecture available in the GrowthOS package.