D
@DBredvick
This is why I’m such a fan of making people do the job manually before they write code to automate it
12
0
506
@levieposted on X
Noticing an interesting version of gell-man amnesia where people use AI for their job and see all the various things they have to do in the “last mile”, but then look at someone else’s job and think that AI will eliminate it immediately. We all have a much deeper appreciation for the nuances and complexities of the work that we do every day. We run into issues about accessing data, we know how much context is needed to get AI models to work the way we need, we have to review the output of the AI to make sure it’s accurate, and then we have to incorporate that work into some broader business process. We see all those steps deeply for the work that we do. Then, a moment later, we see AI do something in a foreign space and think that it can go automate that entire function. We tend to dramatically underestimate the work that goes into making the AI work just as effectively in those jobs. This is reason to be skeptical about many of the theories of job loss. It’s coming from the lens of being able to automate individual tasks with AI, without understanding all the work that goes into doing the job fully.
View original tweet on X →
Community Sentiment Analysis
Real-time analysis of public opinion and engagement
Sentiment Distribution
91% Engaged
82% Positive
Positive
82%
Negative
9%
Neutral
9%
Key Takeaways
What the community is saying — both sides
Supporting
1
"Last mile"
is where the work really lives — a job's hundreds of invisible micro-decisions, edge cases and exceptions that outsiders never see.
2
least familiar with those roles
; the gap between "I saw a demo" and "I do this every day" is where predictions fail.
3
messy data, judgment, accountability, regulation, relationships, and workflow handoffs
that demos gloss over.
4
causes catastrophic errors
, so human judgment and QC remain essential.
5
document and capture the last mile
(often by hiring someone to own it or forcing people to do the job manually first); like outsourcing, the easy 70% gets automated but the hard 30% still needs honest, local handling.
6
more work per worker and more work overall
because cost per output falls.
7
may take years
.
8
consider a range of scenarios, evaluate implications, and map interventions
so policy and practice can steer outcomes productively.
Opposing
1
"good enough" automation
to cut costs — fire a chunk of people and dump the messy leftovers on whoever’s still there, meaning the last mile doesn’t protect jobs, it just concentrates the hardest work onto fewer, lower-paid employees.
2
snapshot bias
AI will keep improving and likely take over last-mile tasks, transforming human roles into opinionated goal-setters and overseers rather than hands-on executors.
3
opposite
suggesting either that last-mile work will continue to preserve jobs or that AI progress will stall — a terse challenge to both dominant claims.
Top Reactions
Most popular replies, ranked by engagement
B
@brianjoseff
the last mile is always where the complexity hides
2
0
150
C
@curtismakes
the people most confident about AI replacing other jobs are usually the ones least familiar with what those jobs actually involve. the gap between "I saw a demo" and "I do this every day" is where every prediction breaks
1
0
178
J
@jessicamalnik
Cute theory. Companies don’t need perfect automation to cut jobs. They just need “good enough” to fire a chunk of people and dump the messy leftovers on whoever’s still there making $20/hour. The last mile doesn’t protect jobs. It just becomes messier problems for fewer people.
0
0
78
Y
@yatterbog
Snapshot bias: Your unspoken assumption is that AI won’t get any better, instead of continued massive improvements that will likely allow it do all of the last mile tasks. People’s roles will be opinionated goal setters.
0
0
56
_
@__smiz
Am I the only one who is the opposite?
0
0
13
Report an Issue
Found something wrong with this article? Let us know and we'll look into it.