Only 13% of 3,362 automation case studies use AI; agents appear in just 18
Sonia_here89 · reddit · 2026-10-19
A Redditor compiled a database of 3,362 vendor-published automation case studies: only 436 (13%) involve any AI, 84 use generative AI, and just 18 feature AI agents (7 in customer support). Key takeaways:
- Most cases are plain plumbing — connecting CRMs to accounting, ERP rollouts, scheduling tools
- AI clusters in support (41% of cases) and invoice/document processing (25%), barely present in payroll, scheduling, or warehouse work (3-4%)
- AI cases don't report better outcomes: time savings 61% vs 58% for non-AI, and cost savings are actually lower (29% vs 36%)
- The model typically just reads messy input; the hours saved come from the boring redesign around it
The author invokes Paul David's 1990 electrification analogy: factories swapped one big motor for the steam engine but kept old layouts, and productivity barely moved for decades until floors were redesigned around the work. Most companies are in the "one-big-motor" phase with AI, echoing MIT's finding that 95% of enterprise genAI pilots show no P&L impact.
Caveats acknowledged: vendor self-reported successes only, a skewed sample, dates skew 2015-2020 so genAI is likely undercounted.
More from AGI Musings
- Hot take: AI guardrails are just content warnings pretending to be intelligence — PierceLilholt · 2026-10-19
- "AI is bad because NFTs" argument draws backlash: every new tech will be compared to NFTs forever — max_paperclips · 2026-10-19
- Compute wins the AI race: Meta caught up and xAI scaled fast by building compute — haider1 · 2026-10-19
- US retailers adopt AI-assisted pricing tools as consumer apprehension grows — ArtificialOther · 2026-10-19
- David Patterson predicts superintelligence will cure all diseases by 2030 — davidpattersonx · 2026-10-19
- GDC 2026 survey: 64% of game devs skip genAI as negative views hit 52% — IanArawjo · 2026-10-19