- Published
Mohammed Ashim has been cutting hair and trimming beards for four years in Bengaluru.
But recently the 22-year old has been strapping a camera to his head to record his work.
asserted
old → publish → work
The camera only records his hands and fingers - his client's faces are blurred out.
asserted
faces → record → hands
"The routine is simple, but you have to be disciplined about how you move.
asserted
you → have → ?
The operations team taught me that my hands always have to stay cleanly inside the frame, the lighting has to be sharp, and I can't tilt my head too abruptly while working."
asserted
I → teach → head
He wears the camera for five or six hours a day.
asserted
He → wear → hours
It's good money, but honestly, the feeling of doing something this futuristic is what keeps me excited.
asserted
what → do → me
I never imagined my razor techniques would be useful to a global tech company," says Ashim.
asserted
Ashim → imagine → company
He is one of hundreds of thousands of workers across the world paid to record their activities.
asserted
He → pay → activities
Ashim's footage is collected by a local company, which might have hundreds of people filming their work.
uncertain
hundreds → collect → work
That video is then sold on to technology companies and research organisations who use it to train artificial intelligence - in particular so-called physical AI, a type of artificial intelligence that can control robots in the real world.
asserted
that → sell → world
Rashdan Technologies collects and processes such human generated data.
"
asserted
Technologies → collect → data
Every robotics company, every frontier AI lab, has a specific kind of data they want: industrial tasks, agriculture, household chores, small business work like pottery, garment-making, embroidery, candle-making.
asserted
they → have → pottery
We go wherever the task is," says co-founder and CEO Shahil Baig.
asserted
Baig → go → ?
His company pays $5 to $10 (£3.74; £7.47) for eight to 10 hours of recorded work.
asserted
company → pay → work
"For a lot of these people, that's meaningful," Baig says.
asserted
Baig → say → people
Making sure workers give their consent and are comfortable with the process is important, he says.
asserted
he → make → process
"We record hands, not faces, and where multi-camera rigs do capture more of the body or environment, we blur faces out.
asserted
we → record → faces
No names, no contact details, nothing that identifies them personally goes into what we hand off.
asserted
we → identify → what
But his customers are getting more demanding.
asserted
customers → get → ?
"Now clients are asking for four to six cameras rigged around a person's head to cover the entire scene, not just the hands.
asserted
clients → ask → scene
Getting someone to agree to one small camera is one conversation.
asserted
someone → get → camera
Getting them to agree to wearing what's effectively a five-camera headset is a completely different, and harder, one.
asserted
what → get → ?
We pay more for that tier of data collection because of what we're asking of people.
asserted
we → pay → people
"
And the industry is becoming more crowded, Agarwal expects dozens more competitors to emerge in India this year.
asserted
competitors → become → India
If you want robots to navigate the real world, they can't just be trained inside computer simulations, argues Manish Agarwal, the founder of Humyn Labs, which collects and processes human-generated data.
asserted
which → want → data
Most of that data comes from video, but sometimes collectors wear hardware on their hands and fingers to capture physical contact.
asserted
collectors → come → contact
"Physical tasks require high-quality signals involving limb movements, fingers, wrists, elbows, shoulders and dynamic force application.
asserted
tasks → require → movements
For example: applying the correct grip force so a plastic bottle doesn't crumble versus a steel glass," he says.
asserted
he → apply → glass
"There is no existing internet archive of embodied human activity - no dataset of a kitchen being cleaned, or a car being assembled by hands that know exactly how much force to apply.
asserted
that → be → force
That data has to be captured from scratch, in the real world, across diverse environments," says Agarwal, whose company has bases in Bangalore and San Francisco.
asserted
company → have → Bangalore
At the moment his firm has 96,000 data collectors across countries in Latin America and Southeast Asia.
asserted
firm → have → America
Humyn Labs filters the thousands of hours of video supplied by those collectors, selecting the most useful content.
asserted
Labs → filter → content
It's then labelled so the AI knows what it is looking at and packaged in a format that can be used by clients.
asserted
that → label → clients
"You need to source across a variety of factories, a variety of tasks, in different geographies because if you want to build deployment for US factories or US houses, you'd want environments that are a proxy for that.
asserted
that → need → that
Residential houses in India may not be the right context for building a robotics use case for the US.
uncertain
houses → build → US
"
But are workers collecting data for machines that will one day do their jobs?
asserted
that → collect → jobs
At the moment it's too early to tell says Bino Paul, senior professor at the School of Management and Labour Studies at the Tata Institute of Social Sciences (TISS) in Mumbai.
asserted
Paul → tell → Mumbai
For some workers AI might be a useful tool, but robots could automate other jobs, he says.
uncertain
he → automate → jobs
The problem is that, if machines do start to take over, there will be little workers can do about.
asserted
workers → start → ?
"By the time workers realise what they have participated in, things may have gone too far," he says.
uncertain
he → realise → what
…and 25 more, not listed.
The Aporia
