In partnership with
If a robot arrived at your workplace, what would you want explained first?
What is a robot-hour? It is the time a business pays to run a robot. Think of it as renting the machine for a shift, rather than buying it outright.
In the US, a business can rent a robot to stack boxes on a pallet for $14 to $18 an hour. The robot is called Cassie, and its maker, Tutor Intelligence, just raised $75 million.
That hourly rate is an eye-catching number. But who teaches the robot? What else is on the bill? And what happens when the skills people use at work become valuable training data?
I am Alex, welcome to ShortCu8 by Innov8.
Lets Dive Deep 🐰
⭐Today's Shortcut
Look at the task: Cassie stacks and unstacks cases. That is one part of a warehouse job, not the whole job.
Read the full bill: The hourly rate comes with a minimum, a setup fee and other upfront costs.
Find the people: Tutor says humans demonstrate tasks and correct robots. Workers whose skills are recorded should know how that data will be used and share fairly in its value.
What does $14 an hour actually buy?
A robot for a particular job: Tutor lists $14 to $18 per robot-hour for palletizing and depalletizing, with a minimum of 80 hours a month. Other picking tasks are priced differently. Tutor pricing
A practical example: At the lowest listed rate, 160 billable hours would cost $2,240 for one month. At $18, the same hours would cost $2,880.
Then comes setup: Tutor lists a $10,000 one-time flat implementation fee, plus $1,000 to $3,500 for freight and $50 to $300 per line for readiness components.
There is also an $8,000 refundable deposit. It is money the customer needs upfront, but it is not the same thing as a fee. Tutor says the hourly rate includes support, maintenance and model updates.
Can we compare that to a salary? Not directly. A robot-hour and a worker's paid hour may cover different amounts of work, and a useful comparison needs the remaining human work and setup costs too.
Who teaches the robot?
People demonstrate the work. Tutor says its human tutors remotely guide robots through tasks so the robots can learn from examples. Tutor research
People correct mistakes. The company says tutors intervene when robots encounter errors. Those demonstrations and corrections become data for improving its models.
People still keep it moving. Tutor says its remote team watches the robots and its technicians handle repairs. The customer’s team supplies empty pallets and takes the full ones away. The company does not say how much human time each deployment needs. Tutor pricing
When your work becomes training data
Meet Ashish Narayan. The Indian Express reported that the Nagpur textile technician wore a camera at work as he adjusted looms and fixed jams. He worried that the recordings could eventually help automate his work, and said he did not know where the footage went. The Indian Express
A separate factory, a separate data collector: The same report described a Tamil Nadu factory whose workers recorded hand movements. The factory had a contract with Objectways, a company that collects and sells data to robotics firms.
Keep the connection honest: The report does not establish that Narayan's footage went to Objectways, or that any of those recordings went to Tutor. These are separate examples of a bigger question: whose knowledge is being recorded, and who gets to benefit from it?
Our take: If someone's skills are being recorded to train these systems, they deserve a clear explanation of how that material will be used and a fair share of the value it helps create.
Does a robot mean a job disappears?
A task can change before a whole job does. Stacking boxes is something a robot can be hired to do; that alone does not tell us what happens to every worker on a warehouse floor.
New roles are not an automatic handoff. Someone may need to supervise, maintain or teach a system. There is no guarantee that the person whose old task changed will get that job.
We do not have evidence that Tutor's deployments caused layoffs. The question is worth asking without inventing an answer.
Before calling a robot “cheaper labour,” look at the whole invoice and the people whose work helps these machines improve. If a worker's hands help teach a machine, that worker deserves to know what their knowledge is being used to build.
The ShortList
🛠️Cool Tools of the Week:
Hydrogen-1.0: Deep Principle's first open-weight model for scientific research
Super-Agent: DeepSeek launches new open-source AI coding agent prototype
Claude: Message threads are now available on the Claude Platform
📩 Innathe Shortcu8 engane undarunnu 👇️?
We read every reply - just reply to this email and let us know how we can improve !
Appo adutha Shortcu8il kanaam bie…👋
If you read till here, you might find this interesting
#AD1
A Single Typo Shouldn’t Cost You a Sale
A shopper types “running shos” and gets nothing back. Another types “red waterprof trail runners under $150” and hits the same wall. Both knew what they wanted.
Traditional site search relies on exact keyword matching, leaving shoppers stranded at “0 results found” screens.
Gladly understands what shoppers mean, typos and all. Instead of a dead end, they get products that fit what they asked for and a conversation that keeps them shopping.
#AD2
Most teams add a second database for analytics. Then manage sync, lag, and drift forever. TimescaleDB extends Postgres instead. Hypertables, 95% compression, aggregates. No pipeline.