Elon
Musk
became
the
butt
of
more
than
a
few
jokes
after
internet
users
pointed
out
Tesla’s
robot
demo
wasn’t
all
it
appeared
to
be.
As
it
turns
out,
a
video
the
billionaire
posted
of
Optimus,
the
company’s
much-hyped
humanoid
robot,
was
actually
being
controlled
by
a
human
slightly
off-screen.
And
it’s
interesting
to
see
robot
manufacturers
now
include
assurances
in
their
videos
that
they’re
not
doing
the
same
deceptive
magic
trick
as
Musk.
First, a quick lesson in recent history if you’re not familiar with the story. Musk has been hyping up Optimus recently, pledging that Tesla would eventually deliver an amazing new robot that people would buy in stores. He first announced his robot in the summer of 2021, but it was just someone literally dressed in a robot costume.
Musk
often
posts
videos
of
Optimus,
but
they’ve
been
underwhelming,
to
say
the
least.
Finally,
when
Musk
posted
a
video
back
in
January
of
Optimus
folding
a
shirt,
eagle-eyed
viewers
noticed
a
hand
that
kept
slipping
into
frame,
clearly
showing
someone
was
actually
operating
the
robot.
A
video
released
by
Elon
Musk
in
January,
with
a
red
arrow
annotation
showing
the
teleoperator
of
the
robot
Gif:
Tesla
/
Gizmodo
The technique here is called “teleoperation,” and has been used in robotics since the 1940s. Essentially someone moves their own hand and the robot mimics the movement. It’s cool for mid-20th-century tech, but it’s not the kind of autonomous robot movements that people here in the 21st century expect for cutting-edge and futuristic products.
And
all
of
that
brings
us
to
an
interesting
phenomenon
we’re
starting
to
see
in
the
wake
of
Musk
getting
embarrassed
by
his
robot
fakery.
Robot
companies
are
now
including
notices
when
they
post
new
demo
videos
that
make
it
clear
the
machine
is
operating
autonomously.
One
example
is
a
new
video
from
Chinese
robot
maker
Astribot.
The
company
posted
a
new
video
this
week,
available
on
YouTube,
showing
the
Astribot
S1
doing
a
number
of
tasks,
including
everything
from
pouring
a
glass
of
wine
to
ironing
a
shirt.
The
robot
can
even
pull
a
tablecloth
from
underneath
a
stack
of
wine
glasses,
a
trick
we
all
half-expect
to
fail
spectacularly.
Astribot S1: Hello World!
The Astribot S1 even folds a shirt in the new video, just like Optimus, but you’ll notice something really interesting in the lower left-hand corner. Those words, “no teleoperation,” probably wouldn’t have been necessary before Musk tried to pull a fast one back in January. But now, as you can see below, it’s a way for robot companies to reassure viewers their robot is actually doing something autonomously without an invisible human hand guiding the process.
Screenshot: Astribot /YouTube
And it’s not just Astribot. The robot company Figure, which uses OpenAI software for its vision software, recently made clear it wasn’t using teleoperation, or teleop, in a very impressive demo released in March.
Figure co-founder Brett Adcock explained the video on X, “The video is showing end-to-end neural networks. There is no teleop. Also, this was filmed at 1.0x speed and shot continuously.”
Canadian robotics company Sanctuary AI released a new video in April that also included a slate explaining that its robot was “autonomous,” reassuring viewers there wasn’t any weird teleoperating puppetry at work.
Sanctuary AI - Phoenix at Human-Equivalent Speed
Musk has a long way to go to catch up to the most innovative robot companies like Boston Dynamics, which just recently retired the hydraulic version of its robot Atlas to devote time to an electric version. But at least he helped provide a public service by increasing transparency in the robotics space.
Nobody
wants
to
get
caught
fudging
a
demo.
That’s
the
kind
of
thing
that
makes
people
incredibly
reluctant
to
trust
you
in
the
future.
At
least
when
Musk
brought
out
a
human
dancing
in
a
robot
suit
everybody
knew
it
was
fake.
When you subscribe to the blog, we will send you an e-mail when there are new updates on the site so you wouldn't miss them.
Comments