Generative
AI
can’t
replace
lawyers
yet
—
unless
we
count
getting
sanctioned
as
a
core
legal
skill
—
but
maybe
it
can
handle
some
of
those
high-end
paralegal
duties.
Like
the
ones
you
delegate
to
your
$2000/hr
paralegals!
Wait,
what?
“Lawyers”
are
telling
her
this.
Look,
inflation
was
bad
for
a
few
months
there
but
it
was
never
“paralegals
billed
out
at
$2000/hr”
bad.
Even
if
we
generously
assume
the
lawyers
were
talking
about
Biglaw
junior
associates
instead
of
paralegals,
they’re
still
topping
out
at
around
$1000/hr.
But
this
story
also
doesn’t
make
any
sense
from
an
industry
perspective.
The
lawyer
creating
that
brief
wouldn’t
be
paying
“$1000-2000/hr”
for
whatever
rough
draft
or
cite
check
she’s
describing,
they’d
be
billing
that
to
the
client.
So
even
if
we
believe
this
completely
implausible
example
where
a
lawyer
is
paying
$60
to
replace
a
$2000/hr
billable…
that’s
going
to
be
a
reason
for
lawyers
NOT
to
adopt
AI.
If
the
firm’s
billable
revenue
significantly
outpaces
its
costs
there’s
no
incentive
to
shift
and
if
paralegals
brought
in
$2000/hr
—
which
is
$4
million
a
year
at
100%
utilization
(which
won’t
happen,
but
you
see
where
this
is
going)
—
there
would
never
be
a
reason
to
automate.
The
example
only
makes
sense
in
the
more
realistic
world
where
the
lawyer
bills
out
the
paralegal
for
$200/hr
but,
with
compensation
and
benefits,
ends
up
underwater
on
paralegal
work
by
the
end
of
the
year
and
GenAI
allows
the
firm
to
eliminate
a
full-time
employee.
And
they’re
not
probably
not
going
to
eliminate
a
full-time
employee
because
no
matter
how
many
tasks
GenAI
can
take
off
their
plates,
the
underlying
models
are
not
going
to
replace
the
most
human
of
tasks
meaning
—
especially
at
the
high
price
points
this
example
supposes
—
the
firm
is
just
forfeiting
a
lot
of
revenue
without
recouping
cost.
All
of
which
is
why
one
generative
AI’s
biggest
impacts
on
the
legal
industry
will
be
the
transition
away
from
the
billable
hour.
Yeah,
yeah,
I
know…
we’ve
heard
that
before.
But
this
time
the
ethical
rules
governing
lawyers
are
going
to
grease
the
wheels
toward
fixed
fee
billing.
Maybe
no
one
is
billing
out
the
paralegal
at
$2000/hr,
but
whatever
the
rate
there
are
still
hours
and
hours
of
paralegal
and
associate
work
that
AI
will
eliminate
through
speed
and
efficiency.
And
what
then?
Is
the
matter
really
the
sum
total
of
hours
worked
on
it
at
every
level,
or
the
value
that
the
senior
lawyer’s
professional
judgment
and
team
management
provides
through
a
finished
product?
If
the
winning
brief
took
300
hours
or
800
hours
shouldn’t
change
the
value
to
the
client.
The
hour,
on
its
own,
is
meaningless…
its
only
value
to
the
industry
is
as
a
crude
estimation
of
the
worth
of
the
final
product.
Something
that
takes
longer
to
prepare,
we
all
assume,
reflected
a
more
complicated
problem
to
solve
or
a
more
refined
final
output
or
both.
But
if
technology
gets
the
same
result
in
half
the
time,
the
hourly
model
can’t
capture
that.
And
since
professional
responsibility
doesn’t
condone
inventing
hours
to
“estimate”
the
time
saved,
the
only
two
options
under
the
rules
for
integrating
these
efficiencies
into
the
financial
model
are
massively
exploding
hourly
rates
or
adopting
a
fixed
fee
for
certain
critical
tasks
that
can
be
based
on
value
rather
than
hours.
So…
maybe
Friar
has
an
unintentional
point
here.
Unless
you
want
to
start
charging
$2000/hr
for
paralegals,
it
might
be
time
to
consider
alternate
or
at
least
hybrid
fee
models.
Joe
Patrice is
a
senior
editor
at
Above
the
Law
and
co-host
of
Thinking
Like
A
Lawyer.
Feel
free
to email
any
tips,
questions,
or
comments.
Follow
him
on Twitter or
Bluesky
if
you’re
interested
in
law,
politics,
and
a
healthy
dose
of
college
sports
news.
Joe
also
serves
as
a
Managing
Director
at
RPN
Executive
Search.