AI Consultant Reflects: Human-in-the-loop Often Just Means Automation Isn't Ready
Warm-Reaction-456 · reddit · 2026-08-01
An AI consultant with 8 years of experience reflects on the industry's popular term "Human-in-the-loop." He argues it's often just a fancy way of saying "our automation doesn't fully work yet."
He used to compare it to autopilot in planes, but realized that in most projects, humans are just reviewing everything the system produces daily—a tedious desk job he invented and billed for. Reviewing past projects, he found effective human interventions were rare, mostly guarding irreversible actions (like refunds or outbound messages). In one "intelligent" content system, the client edited 9 out of 10 drafts, meaning they hadn't automated writing, just the blank page.
He concludes that real loops shrink as systems mature, while fake ones stay the same size forever. He now suggests asking one question at every quarterly review: "What did the human actually catch last month?" If the answer is nothing, either the machine needs more rope, or the reviewer stopped reading months ago.
More from AGI Musings
- Frontier Model Oversight Needs a Dual Track: High Cyber Risks vs. Slow Work Disruption — _akpiper · 2026-08-01
- Berkeley's Agentic AI Summit 2026 Focuses on Ecosystem and Security — ben_burtenshaw · 2026-08-01
- AI Safety Researcher Slams Cynicism Dismissing Rogue AIs as Marketing Stunts — peterwildeford · 2026-08-01
- Dario Amodei Expects AI to Double Human Lifespan in 5–10 Years — PeterDiamandis · 2026-08-01
- Shift Up Faces Backlash for Using AI in Music Video: Is the Hate Fair? — Lucas_Zxc2833 · 2026-08-01
- Why Hire 1000 Art Students When You Can Have Unlimited Michelangelos? — robleclerc · 2026-08-01