What's Dragging Down AI Efficiency? The 'Verification Tax.'
nytimes.com
2 threads
From the article:
“ Call it a “verification tax,” “botsitting” or “rework” — it’s the often arduous task of checking that A.I. output is accurate, safe and usable, and it is putting a very human bottleneck on A.I. automation.”
No “human bottleneck”. A probabilistic technology was never able to automate (this has been a marketing lie for a few years now, old news)
Some automation of human work was deployed long before the massive embrace of LLMs. But managing an LLM automation process should be viewed as a tax - it is a normal process resembling on-boarding a new employee. This is an opportunity to verify that the new 'employee' is doing things the way you want them to be done.