AI Is Coming for This Age-Old
Industry
Many
business leaders don’t yet appreciate the speed at which AI is progressing—and
how rapidly it’s moving both downstream and upstream in the labor task stack.
EXPERT OPINION BY HOWARD TULLMAN, GENERAL MANAGING PARTNER, G2T3V
AND CHICAGO HIGH TECH INVESTORS @HOWARDTULLMAN1
Jan 5, 2026
I spoke recently at an
annual meeting for owners and managers of large commercial construction firms
and, of course, a great deal of the conversation was about the impacts of
automation, robotics and AI on their industry. I came away from the meeting
and the many side conversations with one overwhelming impression—these folks
think that their industry will be one of the very last to be adversely impacted
by AI because, as one guy put it, “software can’t swing a sledgehammer.” The
necessity for large quantities of manual labor in all of their projects would
be their salvation for the foreseeable future.
They had all seen charts
projecting the relative degrees of exposure that various industries had to new
technologies, and typically construction was at or near the bottom of the
lists. They admitted that every few months they were seeing new tools, equipment,
and computer-driven machines being introduced which augmented the abilities of
their workers and took over certain manual functions, but they didn’t envision
a time any time soon when those folks would actually be entirely replaced.
Watching a mobile robot with an extendable arm equipped with a nail gun handle
an entire ceiling of precise installations instead of some poor guy with a
ladder, bursitis and a sore shoulder trying to do the same overhead job, but
taking five times as long, tells you everything you need to know about where
we’re ultimately headed.
My audience members were
actually far more concerned about the fact that—in order to keep up with the
growing national demand—and given the fact that their workforce was rapidly
aging out, the construction industry will need more than 500,000 additional workers
each year. I’ve repeatedly stressed the need for massive increases in vocational training starting in high school and,
in many cases, entirely in lieu of an expensive, debt-infused and
time-consuming traditional four-year college education.
It’s possible that the
industry leaders are finally waking up to the fact that they can’t simply leave
this issue up to the educators or frankly the government if they want something
substantial and timely to be done about the labor shortages. Education is a
business that’s too important and too valuable to be left in the hands of
educators. Of course, because almost everything Trump does largely screws his
own supporters, thousands of skilled “foreign” workers are being seized,
arrested, deported or simply scared off these very job sites across the country
by the masked ICE clowns, so the current labor shortages are getting worse
every day.
I think the unfortunate
aspect of this situation is that the target industries and operators don’t yet
appreciate the speed at which the level of AI understanding and intelligence is
progressing and how rapidly it’s moving both downstream and upstream in the
labor task stack. I have some particular expertise and experience in this
process because I watched a very similar progression in the auto insurance
industry starting decades ago where computer-created estimates quickly replaced
the manual work product of thousands of experienced adjusters.
The computers knew in
minutes the work typically required for a given repair, the parts needed, the
time required to perform the operations, and, of course, exactly what the total
costs of the entire claim should be. If the damage was severe and the car was a
total loss, the computers could calculate and make a settlement offer
instantly. Veteran adjusters with years and years of field work were reduced to
glorified picture takers of the wrecked vehicles involved. With the advent of
the cell phone and the ability of computers to now read and interpret images,
the claimants and insured themselves can submit photos and save everyone the
time and costs of having adjusters travel to physically inspect the cars. When
you add to this situation, the cumulative experience which all the computers
and AI systems now have of thousands of prior wrecks and repairs based on
similar, if not identical, vehicles, it’s possible for the systems to generate
claim settlement offers on the fly and thereby save all of the parties weeks of
costs and delays in resolving accident and theft claims.
As I watch the costing
processes on construction sites these days, the job cost “estimators” whose
expertise is doing take-offs and other calculations to arrive at multi-million
dollar cost estimates for entire construction projects, it’s clear that their
days are also numbered, and the precision guesswork that they do will quite
rapidly be supplemented and eventually entirely replaced by AI systems. This is
another situation where their considerable and hard-earned knowledge and
abilities will quickly be devalued in two respects: first, the nature of the
work that they are charged with estimating will be quickly and radically
changed; and second, the AI systems will have better and more immediate access
to all of the costs of the components, materials, labor, etc. in real time as
compared to the old-time estimators’ seat-of-the-pants guesses.
Here again, a simple
example is the programmable robotic painting machine which can now tape and
paint an entire room in a fraction of the prior time required with sufficient
directions from the construction plans to move around doors, windows and other openings
and precisely apply the exact amount of paint required. While an old-time
estimator might know what time it took Joe, the painter, and his assistant to
get the room done, the AI system already knows what it will cost for the robots
to paint the entire building and how long it will take. This isn’t science
fiction or some distant future vision; this is what’s happening on the ground
every day right now. And finally, it takes one or two guys to watch the robots
rather than a dozen guys to paint the place.
As the farmer said to
his horse: “You won’t lose your job to a tractor, but to a horse who learns to
drive a tractor.”
