The known unknowns of 'up and to the right'
Amazon's new robotics strategy, Intrinsic's industry challenge and finding metrics that matter.
If someone is showing you an “up and to the right” chart, it’s not just glossing over something, it’s hiding dimensions of variables and gating factors. Kent Jenkins
Great comment from Kent Jenkins, startup advisor and Circuit Launch mentor, on my last week’s post re acknowledging the unknown risks when planning the path from A to C.
It’s always tempting to cherry pick metrics (or robot videos) that show improvements, yet every startup risks failure through not selecting the right metrics to worry about. John Doerr’s “Measure What Matters” is an excellent primer. Start with the book, not the worksheets, or it’s too easy to take shortcuts.
Similarly, if you follow the Lean Startup methodology, one of the challenges is avoiding the race towards minimum viable product without asking what question you are solving for. Are you building ‘works like’ or ‘looks like’ and why? Minimum or lean does not mean cheap.
One of the reasons I keep going back to mentor at the Creative Destruction Lab is a focus on finding and testing the right metrics for each startup at every stage of their growth. This isn’t just existential for startups.
It also turns out that the US just bombed Shajareh Tayyebeha, not once but twice or more with the loss of 175-180 lives because of artificial stupidity, not intelligence. Shajareh Tayyebeh was an elementary school in Iran and the targets were 7 to 12 year old girls, not military actors. The culprit was stupid software that increased targeting rate, not scary hallucinating AI.
Palantir didn’t unleash generative AI here except in a natural language wrapper. Palantir focused on quantitive metrics that promised greater speed and accuracy through automation. Increasing the sophistication of technology may deliver better metrics but does not replace having a human in the loop at the right time.
In this instance ‘up and to the right’ disguised a lack of oversight leading to cascading errors. In fact, as any one in Human Factors, HCI or HRI could tell you, technology often recreates the very problem that the technology is supposed to be a solution for. It just shifts or ‘offsets’ the problem somewhere out of sight.
Why did the most sophisticated US targeting systems get this so catastrophically wrong? The military (and capitalist society in general) are on a quest to offset subjective or qualitative human labor, which claims some negotiable importance in a task, with objective or quantitive automation.
We are prone to magical thinking that computer data must be better than human data. This is even seen in Moravec’s Paradox, where we associate computing tasks with greater intelligence. The Palantir Maven system used by US forces to selectively target the Shajareh Tayyebeha school was informed by a database more than 12 years out of date, one which labelled the school as part of a military complex next door. In the intervening 12 years, that area of land was sectioned off and the purpose of the buildings completely changed. A simple google maps search showed it clearly as a school. As did many other information trails detailing the change of use that a human would have checked.
A human analyst would not have selected the school as a target. Why didn’t humans have oversight here? Because we are in love with the rapid speed of AI or machine assisted work. Every layer of automation creates dependencies on dependencies and human oversight in the ‘kill chain’ has been obscured or ‘offset’ as the military calls it. This wasn’t an error caused by advanced AI. Claude had nothing to do with it. This was just automated bureaucracy and artificial stupidity at work.
Sources for this rant on the importance of using the right metrics, automation and warfare come from:
And just for the heck of it… “if you were formerly Speaker of the House, do not tweet this out in earnest”:
Want to host an event or reach the robotics community?
Many companies are reaching out about hosting events, outreach or hiring and we have a great community network, of people looking for work, of investors interested in deal flow and of technologists interested in learning about the latest advances.
Would you be interested in hosting the next Bots, Bevs and Devs? Or something customized specifically for your goals? Reach out here!
Amazon’s new robotics acquisitions
Amazon has just acquired 2 robotics companies:
Fauna Robotics, maker of a $50K humanoid robot built for homes and schools
Rivr, a Swiss startup whose stair-climbing robots tackle last-mile doorstep delivery
CB Insights maintain an interactive Amazon strategy map (for subscribers), revealing the extension of Amazon’s focus beyond warehouse automation.
CB Insights also provide a market map identifying 70+ companies across 10 physical AI model categories. We organized these companies into the following groups:
Data & simulation: Synthetic data generation, real-world robot demonstrations, and simulation platforms where robots learn in virtual environments.
Model approaches: Core AI architectures that enable robots to understand their environment (VLMs), generate actions (VLAs), and predict outcomes (world models).
Foundation models: Pre-trained robot intelligence that combines various model architectures for general manipulation and inference, autonomous driving, and multi-robot coordination.
Observability: Platforms that monitor deployed robots and feed real-world data back into training to improve performance.
Frontier Bio on Wefunder
Frontier Bio, one of our past neighbors here at Circuit Launch, just launched on Wefunder. Frontier Bio creates lab-grown human tissues to reduce the worldwide organ shortage. Their first clinical product is a tissue-engineered vascular graft made from the patient’s own cells. Check out what they are doing with stem cells and bioprinting here!
Train an AI model to solve dexterous manipulation, with $180K in prizes
Calling all developers, roboticists, and engineers!
If you have what it takes to tackle some of the hardest problems in robotics, there’s still time to sign up for the AI for Industry Challenge. Browse the Discourse forum to find teammates and get started on this global competition brought to us by Intrinsic, Open Robotics, NVIDIA, and Google DeepMind. Industry leaders will review your AI models and select the finalists, with $180K on the line and a chance to deploy your solutions in a physical workcell at Intrinsic’s headquarters.
Learn more in this new video from Mike Likes Robots and register here to participate in the Challenge today!
Impressive dexterity demo from Kyber Labs
Boston Dynamics Spot robot dog for Sale
Good condition - robot is 4 years old, used in classrooms and business meetings. Includes tablet controller (UXV technologies), 2 batteries and charger.
Asking $35k obo
Mike Quan, text msg to: 617-500-7520
Silicon Valley Robotics Investor/Startup Database
Are you a startup fundraising atm? Fill out our “For Investors Eyes Only” sheet.
These investors write checks anywhere from angel and seed stage to $5-50 million.
Feedback from the startups - high quality inbound interest from investors.
Feedback from the investors - incredible depth and range of robotics startups.
If you’re an accredited investor and haven’t already signed up for the spreadsheet - Sign up here for great robotics deal flow!
Robotics News
Inside Project Kobe: Amazon’s plan to build Walmart-style supercenters powered by warehouse robots and AI - Business Insider
Amazon just bought a startup making kid-size humanoid robots - TechCrunch
Google partners with Agile Robots, growing its AI robotics footprint - CNBC
US lawmakers to introduce bill to ban government use of Chinese robots - Reuters
Melania Trump pitches robots as potential educators for American schoolchildren - CBS
A Day in Class With Plato, the Melania Trump-Mandated Robot Teacher - The Atlantic
Smarter, Faster, and More Human: A Leap Toward General-Purpose Robots - Technology.Org
These autonomous robots drive to your electric car — and charge it - SF Chronicle
Containerized Autonomous Kitchen Expands Army Food Modernization to the Field - Department of War
Ultrasound Wristband Translates Muscle “Strings” into Robotic Dexterity - Neuroscience News
Ultralightweight sonar plus AI lets tiny drones navigate like bats - The Conversation
Architecting Autonomy at Scale: Raising Teams Without Creating Dependencies - InfoQ
Send me your 2026 progresss, plans and posts for potential inclusion in Robots&Startups.
Robotics Events
Bots, Bevs & Devs at SVCTE in San Jose
April 1 from 6pm to 8.30pm
Details still tbc but see them first on the SVR Luma Calendar
SVR Leaders Circle in SF
April 6 - by invitation only
If you are CEO, CTO, Founder, Chief Scientist or Senior Researcher in a robotics company this is our peer networking event for robotics leaders. Save the date!
IEEE Hard Tech Venture Summit and Robotics Manufacturing Workshop at SRI International - April 16-17 - use “Csvrobo” for $90 discount
AMBayArea Summit 2026 (from the Association of Manufacturers Bay Area) - Chabot Space & Science Center - April 21
Keynote Speaker Andra Keay on Humanoids, Humans, and the Workforce
Revolution No One Is Talking About
More events:
Robotics Invest - Boston - May 26-27
ICRA 2026 - Vienna, Austria - Jun 01 - Jun 05
Automate
Deep Tech Week - SF - Jun 21 - Jun 25
Robot Fringe Festival - London - Jul 2
Robotics: Science and Systems - Sydney, Australia - Jul 13 - Jul 16
Open Sauce - San Francisco - Jul 17 - Jul 19
ROSCon Global - Toronto - Sep 22 - Sep 24
IROS 2026 - Pittsburgh - Sep 27 - Oct 1
IEEE Humanoids 2026 - Silicon Valley - Dec 6-9
The annual IEEE-RAS International Conference on Humanoid Robots is the internationally recognized prime event of the humanoid robotics community. The conference is sponsored by the IEEE Robotics and Automation Society and coordinated by the Technical Committee on Humanoids.
Join the organizing team at Silicon Valley Robotics - bots&beer@svrobo.org
In robotics, the real test is whether the system can withstand real deployment, human oversight, and commercial scrutiny. The sector is moving beyond spectacle and into a phase where execution, governance, and economic viability will matter far more than headline momentum.
Thank you for the mention!