Aug 31, 2026 · 55 min
Industrial automation outpaces the humanoid robot hype
Bullish on Automation and Robotics, but not Humanoid Robots | Shahin Farshchi
The episode reframes robotics as an industrial systems problem, where reliability, economics, and physical training data will determine adoption.
- 1The biggest robotics opportunity lies in specialized automation for factories, warehouses, agriculture, vehicles, and other constrained environments.
- 2Humanoid form factors attract attention, but successful systems will usually combine common components into machines tailored to specific industrial tasks.
- 3Robotics still faces difficult barriers, including reliable performance, convincing unit economics, and far less physical training data than AI chatbots possess.
Don't miss
Farshchi argues that automation could cause short-term displacement while historically producing higher productivity, better-paying work, and broader consumer benefits.
The brief
Shahin Farshchi of Lux Capital argues that robotics is advancing because AI, cheaper manufacturing, and better components now make previously impractical automation viable.
The central distinction is between robots and automation: a useful solution also requires financing, engineering, workflows, software, maintenance, and economics—not merely a machine.
Farshchi sees specialized systems winning across warehouses, vehicle assembly, agriculture, and electronics, while humanoids remain a compelling but often mismatched general-purpose idea.
The conversation’s inflection-point argument compares robotics with computing: standardized building blocks and better platforms could let smaller markets build useful systems.
The hardest problems remain reliability, unit economics, and physical training data; unlike chatbots, robots cannot learn from vast stores of internet text.
The standout tension is economic: automation may displace workers in the short term, yet historically it has also raised productivity, wages, and consumer benefits.
What was said on this episode
34 statements · 20 positive · 9 negative · 3 mixed · 2 neutral
The current hype around robotics is justified.
“Yes, it's absolutely justified.”
Listen at 2:00
Humanoid robots are not the main source of his robotics enthusiasm.
“I'm not particularly excited about the humanoids.”
Listen at 2:26
AI and cheap manufacturing enable complex robotic systems at low prices.
“I'm excited about the most recent wave of innovations in AI and cheap manufacturing and the ability to make things that are extremely complicated, systems that are very complicated at a very low price.”
Listen at 2:30
AI advances and cheaper robotic arms have produced new robotics applications.
“over the past, I would say, 15 years. With AI in the form of convolutional neural nets in the early 2010s and the continued slope of cost reduction with robotic arms, we've seen a whole new wave of new robotics applications.”
Listen at 3:11
Robotics is creating markets that previously did not exist.
“what we're seeing with robotics is that markets for robotics are being created where markets for robots previously didn't exist”
Listen at 7:49
Some robotics companies may dominate newly created markets.
“there's companies out there that are in a position to now dominate those nascent markets.”
Listen at 8:10
Automation includes more than physical robots.
“Robots obviously automate, but automation is not limited to robots.”
Listen at 9:44
Humanoid robots will perform some tasks but remain one of many automation solutions.
“my expectation is that there will be an opportunity. for a humanoid form factor machine that does some subset of tasks that are performed by humans today, but they will be one of many automation solutions that will take on many shapes and forms.”
Listen at 14:40
No single robotic solution will dominate every automation problem.
“I personally, as an end of one, don't see a... single solution dominating all problems”
Listen at 16:05
Specialized robots will generally outperform robots designed for all tasks.
“I generally take issue with the notion of we're building a robot that's going to perform all tasks because I think it'll be very difficult for a robot that performs all tasks to be better than a robot that's specialized to a certain task.”
Listen at 19:01
Robotic engineering is becoming more accessible.
“the engineering that goes into making them useful becoming far more accessible.”
Listen at 21:40
AI interfaces will let users program robots through natural language.
“I see a future where someone can speak to an interface powered by AI that will then program the robot to do a specific task rather than you having to hire an engineer to do that for you.”
Listen at 21:52
Open robotic models can greatly reduce the expertise, cost, and time needed for task-specific training.
“what would have required many PhDs and millions of dollars of funding and many years, now can be achieved by an undergraduate student, maybe even a high school student, downloading and installing one of these models and fine-tuning the robot to perform a specific task.”
Listen at 24:06
Robots still need greater reliability, speed, and labor competitiveness.
“we still have some work to do to make these robots more reliable and to make them faster and to make them competitive. with labor.”
Listen at 24:33
Economies adopting more automation tend to have lower unemployment.
“economies who benefit from more automation, who adopt more automation, tend to also benefit from less unemployment.”
Listen at 28:48
Automation may displace jobs while disproportionately creating higher-quality, more satisfying work.
“when you automate, Yes, you may be displacing jobs, but you are disproportionately creating higher quality jobs where you have less churn, where you have greater worker satisfaction.”
Listen at 29:58
Successful automation will increase selection, competition, and consumer benefits.
“more selection, more competition, and ultimately a benefit to consumers.”
Listen at 32:28
Simple and reliable automation will benefit producers and consumers.
“if you're able to make automation simple enough to adopt and make it reliable enough, then everybody from the people that sell products to these automation customers to the people that purchase the products from them will ultimately benefit.”
Listen at 33:03
Agricultural automation helped shift society from food shortages toward abundance.
“agriculture is one of the first industries to be automated. And as a result of that, we went from food shortages to food being... completely abundant as a result.”
Listen at 34:26
Domestic companion and household robots warrant caution.
“I am a little cautious about those kinds of use cases”
Listen at 35:36
Humanoid domestic robots will arrive later than many expect.
“I think we'll see that come. But I feel like that is further out.”
Listen at 36:57
Domestic robots face challenging adoption conditions.
“I think it would be challenging to introduce a piece of technology like that into the domestic setting.”
Listen at 38:32
Domestic robotics adoption will take longer than many expect.
“it'll take longer than a lot of people think.”
Listen at 38:43
Robotics adoption depends on application breadth, reliability, and demonstrable unit economics.
“It really comes down to the breadth of applications, the reliability, and being able to demonstrate the unit economics to customers.”
Listen at 39:26
It remains uncertain whether scaling alone can solve robotic intelligence.
“the jury is still out as to whether you can, simply solve this problem with scale”
Listen at 46:05
Robotic AI will likely combine language models with traditional sensing and planning.
“it's my expectation that it's going to be some kind of hybrid of the two”
Listen at 46:39
Vision-language-action systems may be slower and less reliable than traditional approaches.
“they may not be as quick and they may not be as reliable.”
Listen at 48:35
Language-based robotic models are more generalizable.
“they're more generalizable.”
Listen at 49:28
Language-based robotic models can learn tasks from demonstrations.
“You can teach them to do things by simply just showing them the task the same way you would show a child a task.”
Listen at 49:30
Robotics companies with application-specific training data will have an advantage.
“companies that are able to access and build these training sets for these particular applications will certainly be advantaged.”
Listen at 51:27
Autonomous cars could independently find parking and leave the user.
“the vehicle could figure out where to park and go off on its own and for me not to have to deal with it.”
Listen at 52:18
Household robots will face adoption barriers until proven safe around children.
“it's extremely unlikely that my wife would allow something like that in the house until it's proven to be safe”
Listen at 53:15
Listeners should learn more about Physical Intelligence.
“They should absolutely learn more about physical intelligence.”
Listen at 53:48
Listeners should learn about Formic and its industrial robot deployments.
“They should also learn about Formic, which is deploying robots in real manufacturing logistics settings.”
Listen at 53:57
Statements are attributed to the speaker as said on the episode and reflect their view at the time, not PodLume's. They are not advice.