The
first
bullet
point
in
any
conversation
about
generative
AI
is
speed.
Check
out
how
fast
it
found
that
key
document,
look
at
how
fast
it
put
together
that
pitch
deck,
observe
how
quickly
it
got
lawyers
sanctioned.
That’s
not
to
say
there
isn’t
a
ton
of
conversation
about
reliability
and
confidentiality,
but
it
boils
down
to
speed.
If
AI
wasn’t
fast,
we
wouldn’t
be
talking
about
it
at
all.
But
the
talk
about
AI’s
speed
always
focuses
on
the
front-end
of
the
workflow.
AI
can
get
that
draft
faster
or
summarize
the
documents
faster,
which
may
be
true,
but
as
a
selling
point
overlooks
how
AI
can
speed
up
the
back-end
by
providing
work
product
that’s
easier
for
lawyers
to
review,
query,
and
edit.
This
oft-neglected
angle
struck
me
during
ILTACON
2024,
as
NetDocuments
showed
me
its
latest
AI-driven
tool,
ndMAX
Assist
—
the
logical
evolution
from
its
PatternBuilder
work.
The
ndMAX
Assist
tool
gives
users
an
intelligent
agent
that
takes
their
natural
queries
to
interact
with
the
documents.
Another
reminder
that
generative
AI’s
most
impactful
use
case
right
now
might
be
in
giving
less
tech
savvy
lawyers
as
smooth
and
comfortable
user
experience
to
entice
them
into
working
directly
with
existing
tools.
Why
ask
an
associate
to
go
into
the
system
and
find
something
when
pinging
the
generative
AI
tool
allows
the
senior
lawyer
to
do
it
directly?
Josh
Baxter,
CEO
of
NetDocuments,
emphasized
that
ndMAX
Assist
creates
“an
easy,
accessible
way
for
legal
organizations
to
introduce
AI-powered
capabilities
into
their
everyday
workflows.”
And
it
does
this
without
lawyers
needing
to
export
content
into
a
third-party
platform,
respecting
the
ethical
walls
law
firms
erect
around
client
data.
So
unlike
ChatGPT
stealing
the
headlines
and
racking
up
sanctions,
ndMAX
Assist
respects
security
protocols
and
operates
in
a
familiar
environment.
Having
Microsoft
Copilot
integration
also
allows
users
to
securely
tap
into
Copilot’s
capabilities
directly
within
the
NetDocuments
environment.
However,
this
is
all
still
about
speeding
up
the
front-end
of
the
lawyer
workflow.
But
during
the
demonstration,
NetDocs
shared
a
client
case
that
landed
a
little
bit
different.
According
to
NetDocs,
Boies
Schiller
recently
hailed
the
company’s
technology
for
helping
get
the
firm
up
to
speed
on
a
major,
discovery
intensive
matter
with
a
tight
deadline.
Using
generative
AI
to
speed
the
summarization
of
50-some
odd
deposition
transcripts
overnight
is
a
classic
example
of
getting
work
fast,
but
what
stood
out
was
an
offhand
remark
about
the
NetDocs
generated
summaries
sped
up
the
review
of
the
work
product.
Traditionally,
this
project
would
require
snagging
10
or
12
associates
who
thought
they
were
about
to
go
home
for
the
weekend
and
telling
them
to
start
diving
into
dense
transcripts
about
a
case
based
on
a
30,000-foot
description
with
only
the
benefit
of
the
slice
of
discovery
that
they’re
assigned.
The
record
in
a
long-running
litigation
can
get
more
self-referential
than
the
MCU
and
some
poor
associate
assigned
three
late-stage
transcripts
can
easily
miss
context
and
nuance
buried
in
earlier
depositions.
By
extension,
this
leads
to
a
jumble
of
inconsistent
write-ups
full
of
irrelevant
details
or
ultimately
abandoned
theories.
And
that’s
before
considering
that
some
associates
just
write
better,
more
consumable
summaries
than
others.
This
variability
forces
a
partner
to
spend
the
weekend
wading
through
all
of
it
—
just
to
figure
out
what’s
actually
relevant.
It’s
billable
legal
work,
sure,
but
it
also
sucks.
Interrogating
the
discovery
set
through
a
tool
like
ndMAX
Assist
will
produce
consistent
memos,
produced
in
the
same
voice,
informed
by
the
entire
universe
of
material.
Now
there’s
no
rogue
insights
or
unnecessary
tangents
and
senior
lawyers
trying
to
get
a
handle
on
what’s
in
each
of
these
transcripts
and
what
needs
to
be
revisited
for
further
focus
can
swiftly
push
through
the
process
with
a
set
of
memos
in
the
same
voice
with
the
same
level
of
detail.
BSF
partner
David
Simons
confirmed
the
story,
explaining
that
the
team
was
impressed
with
the
back-end
accuracy
of
the
NetDocs
generated
summaries
and
“having
these
in
a
consistent
voice…
allowing
apple
to
apple
comparisons
without
worrying
about
different
writing
styles
or
technical
abilities.”
AI
skeptics
worried
about
asking
an
AI
to
produce
these
summaries
miss
the
whole
point.
This
isn’t
turning
over
legal
work
to
a
machine
because
no
one
is
going
to
take
these
summaries,
crack
their
knuckles,
and
head
into
court.
But
you
also
wouldn’t
trust
a
team
of
first-year
memos
for
that
either.
This
is
about
getting
to
the
next
stage
where
you
understand
who
the
players
are
and
their
relationships
to
each
other
and
how
the
timeline
went
down
and
it’s
all
not-very-intellectually
demanding
stuff
that
AI
can
deliver
faster
and
in
a
form
that
lawyers
can
work
with
faster.
They’re
on
to
the
stage
of
tasking
associates
to
dig
deeper
into
specific
transcripts
or
begin
researching
key
concepts
within
hours
instead
of
days.
“At
the
end
of
the
day,
in
a
situation
like
that
where
you
want
to
do
the
same
thing
50
times,
asking
the
same
thing,
getting
the
same
meticulously
approved
prompt,
or
series
of
prompts
that
are
strung
together
to
get
you
an
outcome
that
you
feel
good
about,
that’s
important.
And
that’s
something
that
you
have
to
have
a
tool
like
ndMAX
to
do,”
Chief
Product
Officer
Dan
Hauck
said.
No
one
expects
front-end
speed
to
disappear
from
AI
sales
pitches.
Turning
a
100-hour
project
into
a
couple
hours
will
always
be
the
sexier
headline
for
firms
and
(more
so)
clients.
But
don’t
blow
off
the
downstream
workflow
advantages
either.
Legal
work,
in
particular
litigation,
is
an
iterative
process
meaning
the
faster
lawyers
can
move
from
absorbing
new
information
to
formulating
the
next
research
task
is
just
as
important
as
turbocharging
the
process
of
accumulating
the
information.
Time
—
especially
billable
time
—
is
fungible
after
all.
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.