Michael May

Interesting comparison of human-based ride hailing services (UberX, Lyft) vs autonomous-driving one from Waymo. TLDR: Waymo is more expensive and a lot slower to get you to your destination. And this is in the city where Waymo vehicles and software are probably most optimized for. This isn't to bash on Waymo and other self-driving car platforms, which have made incredible progress. That being said, the problem is not solved and it will take many more billions and many more years to get autonomous driving services like this to be price and performance competitive to current ride hailing services. https://lnkd.in/eb4uZ2Um #autonomouscars #AI

2

The future of the automotive supply chain is here! 🚀 At CES 2025 today, the secret is out. 🌍🔧 Game changing! Find out more at: https://supplywhy.ai #CES2025 #AI #GenAI #AutomotiveTech #SupplyChainInnovation #SaaS

10

1 Comment

Watch this funny video of a Waymo getting stuck in a doom loop driving endlessly around a roundabout. AVs still have some kinks to resolve. “You have to wonder where the teleoperators were during this meltdown. If you weren't aware, robotaxi companies like Waymo employ round-the-clock teams of remote technicians that take over vehicles when they get stuck or go haywire. Maybe they weren't alerted of the issue, or maybe it genuinely took them some effort to wrest control back over the robotaxi.” Wonder indeed… why didn’t the teleoperations step in sooner here? Upshift is launching a teleoperated car subscription service in January. We won’t drive you in circles! Subscribe to a car that comes to you, drives you around, runs your errands, and valet parks itself. Learn more and join the waitlist at https://lnkd.in/gemiY4Sh

6

6 Comments

Waymo grew from 100K rides per week to 150K rides in two months. At 7.8M rides per year this is nothing compared to the 10B rides Uber provides. But here is the catch, Uber's costs are more or less fixed. Google's are going to come down exponentially. The cost for an AI team will go down very fast if expressed per ride. Given no drivers are needed and vehicle costs are coming down as well, it is just a matter of scaling up for Google to steal away clients from first Uber but afterwards also public transport, second car ownership and finally first car ownership. In 2025, Waymo can easily go to 1 or multiple millions of rides per month. 2026, 10-50M and in 2027, 1B per month. Giving that Tesla will want to launch in 2025 and scale in 2026, consumers will be the big winners. https://lnkd.in/e4rcAH6D

13

6 Comments

🚗✨ Step Inside the Future: Waymo's 360° Autonomous Driving Experience Imagine gliding through the streets in a vehicle that drives itself—safely, seamlessly, and autonomously. With Waymo's 360° Experience, you’re not just a passenger; you’re a witness to the cutting-edge technology shaping the future of transportation. Here’s what makes the journey extraordinary: 🔄 360° Awareness The Waymo Driver uses advanced sensors and AI to detect everything around the car—pedestrians, cyclists, and even that unexpected ball rolling into the street. 🛣️ Navigating Complex Roads From busy intersections to quiet residential areas, the Waymo Driver handles it all with precision, ensuring a smooth and safe ride every time. 🤖 Learning with Every Mile Every trip adds to Waymo's vast knowledge base, improving its driving intelligence and making your journey even more reliable. 🌟 A Glimpse of Tomorrow This isn’t science fiction—it’s happening now. Fully autonomous driving is transforming mobility for individuals and communities alike, offering accessibility, safety, and convenience like never before. 🚀 Ready to experience the road reimagined? Dive into the details and explore the world of fully autonomous driving here: [Insert Link] #Waymo #AutonomousDriving #Innovation #AI #FutureOfMobility

3

Notion vs Glean reminds me a lot of Tesla vs Waymo... Just like I think Tesla is going to solve the 'software' problem of autonomous vehicles before Waywmo solves the 'hardware' problem... I think that Notion is going to solve 'enterprise knowledge search' before Glean solves the problem of document and knowledge creation... Why? A couple reasons: 1) Just like Tesla, Notion has a huge advantage when it comes to the amount of underlying data and documents to train on... They already have 100M+ users, and likely 10x that in terms of the number of documents, files, etc. in team's workspaces. 2) It's very hard to change people's behaviour re: where/how they create 'work'. It's why a lot of companies have come and gone in that category... As exciting as Glean's AI writing assistants might seem, most people are just fine (or even prefer) a good old fashioned editor/writer. I could be way off here, but this will be an intersting one to follow. Thoughts? Where am I getting it wrong? #ai #enterprise #knowledge #management #notion #glean

3

1 Comment

Waymo's co-CEO taught me something unexpected about market choices: The so-called 'simple' path can lead to a massive dead end. That's why Waymo avoided what many believed was the obvious route — autonomous trucking. It seems like it should be the reverse. Autonomous trucking focuses on highways because there are: • Fewer variables • Less complexity • Straightforward routes • More predictable conditions • Fewer pedestrians, etc It sounds logical, right? But here's the reality of it: Highway accidents at 70mph with 20x the mass are catastrophic. The margin for error is almost zero. And, the trucks still have to eventually navigate into cities to pick up and drop off deliveries. Meanwhile, Waymo took the "harder" route. They focused on urban environments with: • Unpredictable pedestrians • Complex intersections • Endless variables Now it provides over 100,000 paid rides each week. By solving the long tail of possible scenarios the cars can encounter, they’ve catapulted ahead. And it’s just getting started. So what’s the lesson here? Starting with the "harder" problem can be smarter than taking shortcuts when the less hard version is a false proxy for the real thing. It means you: 1. Build robust solutions 2. Tackle core challenges head-on 3. Develop deeper expertise 4. Create transferable technology Waymo's journey since 2009 proves it: • Started in complex urban environments • Built sophisticated systems • Created resilient technology • Achieved real-world success Meanwhile, highway-focused companies are still struggling with the "easier" problem. The takeaway for startups: Don't chase the easy path. Chase the right path. Sometimes, embracing complexity early leads to simpler solutions later on.

17

15 Comments

After almost a decade working on Civil Maps and through the Luminar acquisition, I’m moving on and taking a break! So proud of our fantastic team. I’m confident they’ll get our lidar software in production vehicles 🙂 I’d also like to share my passion project ScreenZen (iOS, Android, MacOS). It helps you reduce your screen time on distracting apps like Instagram. I started it by trying to improve my screen time during the pandemic and has been quite rewarding to share with a larger audience. ⚙ How it works: 1️⃣ Adds friction: Wait before your distracting app opens 2️⃣ Intentionality: Budget your # of app opens for the day 3️⃣ Catch excessive scrolling: A check-in prompt after a certain time on an app 4️⃣ Streaks: People love streaks 📈 Reception: 80k DAU and 140k MAU without ads/marketing. Growth mostly from unpaid influencers sharing it (biggest one was Drew Gooden ~4.5M views) and app store searches 💲 Business Model: Currently free with a tip that unlocks lifetime access if it goes paid in the future. Tip now before it's too late! 😉 Digital Wellness is an exciting, nascent industry that I believe will only grow as tech continues to advance and become more ingrained in our lives. So I'm going to work on ScreenZen, but also really try to take a break after 10 years on the startup grind 👍

152

24 Comments

In my opinion, this is one of the lesser discussed disruptions that we will experience in this decade. While widespread availability and economies of scale are yet to be fully figured out, this has enormous potential to reinvent last mile logistics, transportation and also road safety. Exciting times ahead !! https://lnkd.in/g9iUeADd

16

Tesla’s autonomous cars are making headlines for being… not so autonomous. Remote humans sit behind the scenes to supervise and step in (hopefully they’ve got driving licenses!). Perhaps this approach is driven by regulatory constraints (like with Waymo). Or perhaps the tech is not quite there yet. For me, the issue here is transparency, not the humans. For the near future, for AI to have proper impact on large-scale, real-world problems - where reality is messy and the cost of mistakes is high - we’re going to need to blend it with humans. In fact, the magic is in how to blend them: 1. Precisely calibrating human involvement to ensure high accuracy without wasting effort. 2. Maximizing the system’s ability to learn and improve so humans can step back over time. Getting this balance right isn’t a stopgap—it’s the bridge to the future of autonomy.

67

2 Comments

Sharing some further reflections on AI product analogies, governance and what that entails in an attempt to bridge the point Yann is making, that AI will not magically emerge, we will design and build these so called sophisticated AI systems and therefore it is unrealistic to preemptively regulate it to prevent existential risk. I think, we (designers/builders) should not delay governance and measurement capabilities needed for AI products and iterate on both in tandem. Before policies and regulations can be defined there is a lot of work to be done to solve for tools and methods to observe, gather metrics and develop benchmarks, methods and standards in partnership with orgs like NIST. One other challenge that makes defining governance capabilities for AI products tricky is that we are at an extremely early infancy stage with this technology that has no really good analog. AI is like multiple things simultaneously making comparison to aviation, electricity, bioengineering, education, nuclear weapon … only partially accurate. Like Yann, I have been using the aviation analogy calling these current AI products akin to the glider planes in early 1900s with one nuance, I think in terms of four layers of analogs based on the application, usage and implications. First layer is ground breaking technology that is so magical it captures our wildest imagination like aviation aka gliders to turbo jets, the next layer is systems that profoundly change society eg. education and health care system, third layer is ubiquitous and critical infrastructure technologies like electricity, network, that everyone should be able trust and safely use and finally the fourth layer is adaptive and autonomous technologies where we do not have much of an analog today but task based self-learning autonomous products are emerging, e.g. your social feed or e-commerce recommendations. Governance typically entails transparency, privacy, fairness, observability and explainability and technical robustness. It is nearly impossible to measure explainability and evaluate fairness without the context in which AI product is being used or implemented. So by focusing on the layer(s) at which the AI product is operating via use case and understanding of its consumers, measurement and tracking needs can be defined. Every industry developing AI products should plan for investment needed in such capabilities. This is my perspective so far, let me know your thoughts 😊

8

Please give up already Elon. As the CEO of Tesla, he should at least be aware that CNNs are pre-dominantly used in object detection and FSD. (Or pretty much any other vision task) ViTs are heavier, not as robust as CNNs and performance wise, not as good as CNNs Generally in practice CNNs are used as feature extractors with few attention heads.

186

12 Comments

Caltrain plans to introduce bi-level dual electric and battery-powered train to extend its zero-emission service. These trains will charge their batteries in electrified zones between San Francisco and San Jose and use the stored power for non-electrified segments, such as the extensions to Gilroy. This is the next step in battery-operated vehicles for mass transportation, towards sustainable and eco-friendly transit solutions.#sustainability #transportation

6

Elon Musk Shares Glimpse Into The Near Future Of FSD: One Intervention Per Year I've been using Tesla's FSD version 12.3.3 to 12.3.6 for the last couple of months now and most of the time, it drives very well. According to Elon Musk, a future version of FSD, which I think means 12.6, will have only one intervention per year. The wording here is VERY interesting. An intervention, as what Elon described here, is doing something like hitting the accelerator pedal to encourage the car to move forward. This is different from a disengagement, which shuts off FSD completely and has the driver taking over all driving. I am going to say that Elon Musk means an intervention here and if that is true, is it also true that a disengagement, where the driver has to take over, will only happen once per year - or will it be even less? I think it's going to come down to the data of usage of FSD. Right now, I disengage, but it usually isn't for safety critical reasons. It's for semantics, such as the car not getting in the far left lane soon enough to avoid having to cut cars off to get in that lane down the road to make a turn half a mile away. The semantics of driving haven't been talked about much, but the only way to know if there is an improvement there is to stay subscribed to FSD and to keep using it in my daily life and in new places. I'll be able to see for myself if it improves in its route planning. Full Article: https://lnkd.in/gk-Fwr9r $TSLA #tesla #fsd #autonomy #software #driverless #robotaxi

7

1 Comment

Where is my self-driving car? My 2015 Tesla can drive on highways, maintain speed, and change lanes autonomously. Yet, here we are nearly a decade later, and self-driving cars are still not a reality. This comes down to a basic rule of software, the last 10% of progress requires 90% of the effort. While that 2015 functionality was impressive, and made it feel like self-driving would be any day, it doesn't reflect the full picture. This is worth considering when evaluating whether Large Language Models (LLMs) will replace programmers—or any other profession. The demos, videos, and examples showcasing LLMs are impressive, and easily handle simple tasks. However, they struggle with that final, hardest 10%. Transitioning from being a helpful tool to a full replacement is a massive hurdle.

20

1 Comment