AI In Autonomous Vehicle Technology

I came across a new framework that brings clarity to the messy world of AI agents with a 6-level autonomy hierarchy. While most definitions of AI agents are binary (it either is or isn't), a new framework from Vellum introduces a spectrum of agency that makes far more sense for the current AI landscape. The six levels of agentic behavior provide a clear path from basic to advanced: 𝐋𝐞𝐯𝐞𝐥 0 - 𝐑𝐮𝐥𝐞-𝐁𝐚𝐬𝐞𝐝 𝐖𝐨𝐫𝐤𝐟𝐥𝐨𝐰 (𝐅𝐨𝐥𝐥𝐨𝐰𝐞𝐫) No intelligence—just if-this-then-that logic with no decision-making or adaptation. Examples include Zapier workflows, pipeline schedulers, and scripted bots—useful but rigid systems that break when conditions change. 𝐋𝐞𝐯𝐞𝐥 1 - 𝐁𝐚𝐬𝐢𝐜 𝐑𝐞𝐬𝐩𝐨𝐧𝐝𝐞𝐫 (𝐄𝐱𝐞𝐜𝐮𝐭𝐨𝐫) Shows minimal autonomy—processing inputs, retrieving data, and generating responses based on patterns. The key limitation: no control loop, memory, or iterative reasoning. It's purely reactive, like basic implementations of ChatGPT or Claude. 𝐋𝐞𝐯𝐞𝐥 2 - 𝐔𝐬𝐞 𝐨𝐟 𝐓𝐨𝐨𝐥𝐬 (𝐀𝐜𝐭𝐨𝐫) Not just responding but executing—capable of deciding to call external tools, fetch data, and incorporate results. This is where most current AI applications live, including ChatGPT with plugins or Claude with Function Calling. Still fundamentally reactive without self-correction. 𝐋𝐞𝐯𝐞𝐥 3 - 𝐎𝐛𝐬𝐞𝐫𝐯𝐞, 𝐏𝐥𝐚𝐧, 𝐀𝐜𝐭 (𝐎𝐩𝐞𝐫𝐚𝐭𝐨𝐫) Managing execution by mapping steps, evaluating outputs, and adjusting before moving forward. These systems detect state changes, plan multi-step workflows, and run internal evaluations. Examples like AutoGPT or LangChain agents attempt this, though they still shut down after task completion. 𝐋𝐞𝐯𝐞𝐥 4 - 𝐅𝐮𝐥𝐥𝐲 𝐀𝐮𝐭𝐨𝐧𝐨𝐦𝐨𝐮𝐬 (𝐄𝐱𝐩𝐥𝐨𝐫𝐞𝐫) Behaving like stateful systems that maintain state, trigger actions autonomously, and refine execution in real-time. These agents "watch" multiple streams and execute without constant human intervention. Cognition Labs' Devin and Anthropic's Claude Code aspire to this level, but we're still in the early days, with reliable persistence being the key challenge. 𝐋𝐞𝐯𝐞𝐥 5 - 𝐅𝐮𝐥𝐥𝐲 𝐂𝐫𝐞𝐚𝐭𝐢𝐯𝐞 (𝐈𝐧𝐯𝐞𝐧𝐭𝐨𝐫) Creating its own logic, building tools on the fly, and dynamically composing functions to solve novel problems. We're nowhere near this yet—even the most powerful models (o1, o3, Deepseek R1) still overfit and follow hardcoded heuristics rather than demonstrating true creativity. The framework shows where we are now: production-grade solutions up to Level 2, with most innovation happening at Levels 2-3. This taxonomy helps builders understand what kind of agent they're creating and what capabilities correspond to each level. Full report https://lnkd.in/gZrGb4h7

The future of AI isn't just about what agents can do, it's about how much autonomy we give them. I just read a fascinating research paper from the Washington University in St. Louis that breaks down AI agent autonomy into 5 distinct levels, from L1 (user as operator) to L5 (user as observer). This framework could fundamentally change how we design and deploy agentic AI systems. Here's why this matters for builders and innovators:- 1. L1 and L2 agents keep humans in the loop as collaborators and operators, perfect for skill development and high-stakes decisions where accountability matters. 2. L3 agents act more like consultants, taking the lead but checking in for expertise and preferences. Think Gemini Deep Research or GitHub Copilot Agent. 3. L4 and L5 agents operate with minimal oversight, only seeking approval for risky actions or running fully autonomous with just an emergency stop. The key insight? Autonomy is a deliberate design choice, separate from capability. A highly capable agent can still operate at low autonomy levels when appropriate and sometimes that's exactly what we want. As we build the next generation of AI solutions, we need frameworks like this to help us calibrate the right level of human-AI collaboration for each use case. The question isn't "can AI do this task?" but "what level of autonomy should AI have for this task?" What autonomy level makes most sense for your AI projects? #AI #AgenticAI #AIGovernance #Innovation #TechStrategy #HumanAICollaboration

On my recent trip to San Francisco, I had the chance to experience a Waymo self-driving car, and it felt like stepping into the future. No driver. No human intervention. Just AI quietly taking charge of something we’ve always associated with human reflexes and instincts. At SilverFern Digital we keep a close eye on such breakthrough experiences- studying how products like these function, helps us absorb key learnings and put them to use in our AI-first products and everyday design practice. We broke it down: how is AI able to do this so seamlessly? 🔹 Studying Patterns: Waymo cars don’t just "react." They’ve been trained on millions of miles of driving data, learning the tiniest nuances of human and environmental behavior on the road. 🔹 Building Intelligent Systems: From perception (seeing pedestrians, cyclists, traffic signals) to prediction (anticipating how others might move), every decision is powered by a layered AI brain working in real time. 🔹 Cohesive UX & Trust: The magic isn’t just in the AI. It’s in how that intelligence is communicated back to passengers. Clear displays, intuitive cues, and subtle motions help you trust the car. That’s where UX becomes just as important as AI. This intersection of AI, UX, and automotive design is reshaping not just how cars move, but how we move, work, and live. For me, the ride wasn’t about tech; it was about how natural it felt to let go, to trust, and to experience safety redefined by design. The future of transportation isn’t just autonomous. It’s empathetic, data-driven, and deeply human-centered. As we build more AI-first products, these innovations inspire us to design new-age automotive experiences that push the boundaries of design, technology, and trust. What are your automotive transformation experiences? #AI #UXDesign #FutureOfMobility #SilverfernDesign

Quantum Systems and AI. The Vector AI drone is a hybrid beast. It takes off and lands vertically like a multirotor, then transforms mid-air into a sleek fixed-wing aircraft for long-range reconnaissance. But what truly sets it apart is what’s inside: dual NVIDIA Jetson Orin processors humming with real-time artificial intelligence. These processors enable the drone to identify and track objects autonomously, filter through visual noise, and prioritize threats — all while flying fully autonomously, even in GPS-denied environments. With AI onboard, Vector doesn’t just send back raw data; it delivers actionable intelligence. Whether deployed solo or as part of a coordinated swarm, it adapts to dynamic mission profiles and terrain like a thinking organism in the sky. Meanwhile, the Twister is Quantum’s compact, rugged answer to tactical ISR in tight spaces. It’s small enough to fit in a backpack, but don’t let the size fool you — Twister packs a high-tech punch. Its AI is multi-modal: visual processors scan and analyze landscapes in real-time, while acoustic sensors — guided by onboard machine learning — listen for distant artillery or mortar fire, triangulating their origin with uncanny precision. Twister doesn’t just see; it hears the battlefield. Both systems are designed to reduce operator load. Instead of relying on constant human control, they use their onboard intelligence to fly missions, recognize targets, and adapt to the unexpected. In effect, they transform the operator’s role from pilot to mission commander — making decisions based on insights the drones themselves produce. With Vector and Twister, Quantum Systems is shaping a future where drones are no longer just eyes in the sky — they are thinking, learning, evolving platforms that bring AI directly to the edge of conflict and crisis response. https://lnkd.in/d4P-EgYw

Autonomous Vehicles The AV revolution is underway. Driven by breakthroughs in AI, compute, and simulation, and dramatic cost reduction in sensors and hardware, Robotaxis are being tested in several U.S. cities. Globally there are more than 30 companies piloting/scaling fleets. In the U.S. there are 10 million workers who drive for a living: a) 3.5M truck drivers, b) 2M ride-hailing drivers (Uber, Lyft), c) 1M delivery van drivers (UPS, FedEx, Courier), d) 500k bus drivers (school & transit), e) 400k taxis, and f) 3M drivers in the GIG economy (food delivery) - representing 6.25% of the total workforce. Globally, there are ~400M workers globally that drive for a living. The truck driver or Uber driver replaced by AV is estimated to cut costs per mile by more than half. The implications are massive. In the U.S annually, auto accidents result in 44,000 fatalities, 2.3 million injuries with an economic cost of $350 billion annually (medical, productivity loss, property damage, legal expense). AVs are expected reduce accidents by 90%+. AI on wheels as one analyst labels it, is powered by neural networks, trained on billions of road miles (Waymo alone has logged 100 million with no human driver behind the wheel). Tesla recently launched its pilot program at a price point well below Uber ($4.20 per ride), while Uber itself plans to deploy 20,000 AV (no driver). Bank of America estimates a $1.2 trillion AV spend on robotaxis, logistics, delivery, agriculture, and public transit. This shift could redefine urban design, free up parking, reduce congestion, and accelerate the timeline for traditional auto ownership where more people use AVs on demand vs. owned vehicles. China may lead the race given its demographic urgency and regulatory structure, but the U.S. isn’t far behind. The winners will be OEMs who master software, data, hardware integration, cost-efficient assemblage. Key technology and components are Radar, LiDAR, Camera, Chips, Cockpit to console with nearly 100 companies providing parts, technology and components that has largely evolved beyond traditional auto parts suppliers My most immediate questions/issues related to the advancement of AV include: - Employment, and potential displacement of active drivers - Demand and profitability for the auto OEMs (GM, Ford, Stellantis vs. Tesla)—new car sales, adoption, fleet size, efficiency. - Auto Parts Supplier relevance in a AV transport world - Rental Car Companies (Avis, Hertz, Budget) vs. Robotaxi model - Auto Insurance, premium vs. payout model with fewer accidents and Tesla providing vehicle insurance from their insurance arm The auto sector has underperformed in 2025; credit spreads have widened. Stay tuned, it’s early days.

Autonomous vehicles leveraging advanced AI like Vision Transformers highlight the potential for safer, smarter transportation systems, where real-time decisions driven by enhanced image analysis could redefine how we navigate urban environments and beyond. Vision Transformers (ViTs) utilize attention mechanisms to process diverse visual inputs simultaneously, enhancing the accuracy of object recognition and decision-making in autonomous vehicles. ViTs require substantial investment in R&D, collaborative partnerships, and regulatory alignment to ensure safe and reliable integration. Training technical staff and gaining public trust remain essential steps for widespread adoption, while companies must also address the cost implications to position themselves competitively in a rapidly evolving market. #AI #AutonomousDriving #VisionTransformers #FutureMobility #Transportation

Transportation is the largest employment sector on Earth. Over 1 billion people globally work in roles directly tied to moving people or goods, drivers, operators, couriers, logistics staff. That industry is now facing a seismic shift. At Viva Technology #Paris, I got a hands-on look at Tesla’s new Robotaxi a fully autonomous vehicle with no steering wheel, no pedals, and no driver seat. Just sensors, AI, and minimalism. Here’s what we know: • Tesla plans to unveil the production version on August 8, 2025 • Initial manufacturing is already underway in Texas • Pricing aims to undercut public transport, not just Uber • It will operate via Tesla’s own ride-hailing app • First cities targeted: Austin, San Francisco, Los Angeles • No human driver — full autonomy powered by Tesla's FSD and Dojo AI stack • Global expansion dependent on regulatory approval and real-world test data Tesla isn’t alone. • Waymo (Alphabet) is running autonomous taxis in Phoenix and San Francisco • Cruise (GM) is paused after safety issues but plans to return • Baidu, Inc. and AutoX are already live in parts of China • Uber partnered with Waymo, but their core model faces existential risk The implications are massive: • Driving is the most common job in 29 US states • Millions of Uber, truck, and taxi drivers globally could be replaced • Cities may need to rethink urban infrastructure, licensing, and labor support • Investors will shift focus to platform owners, not fleet operators We’re not talking about a decade from now. We��re talking about product launches this year, pilots already active, and regulators being pushed to move fast. The transportation sector as we know it is approaching a turning point. Are we ready? #AutonomousVehicles #TeslaRobotaxi #FutureOfWork #TransportationDisruption #MobilityTech #AIandJobs #Tesla #Waymo #Cruise #UberFuture #DigitalTransformation #AIInnovation

The Future of Autonomous Vehicles: How GenAI is Accelerating Innovation . The future of fully autonomous vehicles (AVs) is accelerating, thanks to the transformative power of generative AI (GenAI). As highlighted in recent insights from CB Insights, #GenAI is breaking down key barriers that have long delayed the widespread adoption of self-driving #cars . (1) Enhancing In-Car Communication One major advancement is the enhancement of in-car voice assistants. GenAI-powered LLMs are bridging the communication gap between passengers and self-driving cars, evolving from pre-recorded commands to hyper-personalized, natural conversations. Imagine saying, “Let’s go pick up food at my favorite restaurant,��� and your car seamlessly understanding and acting on it—a future that’s already within reach. (2) Reducing Training Costs Training costs are also being slashed through GenAI-simulated environments. These virtual settings allow AV systems to rack up millions of miles driven in a controlled, cost-effective manner, improving safety testing without the need for extensive real-world trials. This innovation is a game-changer for automakers aiming to refine their technology efficiently. (3) Improving Safety and Transparency Safety and transparency are critical for gaining regulatory trust, and GenAI is stepping up here too. By providing clear explanations for driving decisions—moving away from the “black box” approach—LLMs enhance accountability. For instance, a car detecting a pedestrian and explaining its stop decision in plain language builds confidence among regulators and passengers alike. (4) Strategic Partnerships To stay competitive, automakers must partner with automotive AI chip manufacturers capable of supporting local LLM processing. Factors like inference time, energy efficiency, and durability will be key in selecting the right technology partners. Meanwhile, car insurance providers are adapting by developing new risk assessment models, including provisions for cybersecurity threats, potentially collaborating with automotive cybersecurity firms. (5) Transforming Cars into Digital Platforms Looking ahead, GenAI is turning cars into digital platforms with agentic AI features. This opens doors for automakers and AV providers to team up with AI agent developers, creating smarter, more interactive vehicles. The UK AI #startup PhysicsX, nearing a $1 billion valuation, exemplifies this trend, developing advanced AI tools for automotive and #aerospace sectors that could further propel AV #innovation . EmpowerEdge Ventures

NVIDIA and Mercedes-Benz AG-Benz are pushing autonomy in a more realistic direction, less hype, more reasoning. NVIDIA has partnered with Mercedes-Benz to develop a new generation of driverless systems powered by reasoning-based AI, not just pattern recognition. Built on NVIDIA’s DRIVE platform, the focus shifts from “trained reactions” to decision-making: • Handling rare edge cases • Understanding cause and effect • Explaining why a decision was made • Adapting in real time instead of relying only on pre-trained scenarios That distinction matters. Most current autonomous systems perform well in familiar conditions, but struggle when the environment behaves unexpectedly. Reasoning-based AI aims to close that gap. Mercedes-Benz plans to integrate this technology into future vehicles, starting with an initial rollout in the U.S., where regulation allows higher levels of autonomy. Global expansion will follow, only after real-world validation and regulatory approval, not before. Important reality check: This does not mean fully driverless cars tomorrow. Safety validation, legal frameworks, and public trust remain the real bottlenecks. But it does signal where autonomy is actually heading: from scripted behavior → to explainable, adaptive intelligence. The question isn’t if autonomy improves — it’s whether reasoning-based AI can scale safely in the real world. follow and connect Patrick. #NVIDIA #MercedesBenz #AutonomousDriving #AI #AutomotiveTech #FutureOfMobility

I went through this paper on Levels of Autonomy for AI Agents, and it’s one of the better attempts at treating autonomy as a design decision, not a maturity badge. I did a quick exercise mapping autonomy levels to real security tasks. Research Paper > https://lnkd.in/d3p-MvqB ▪️For 𝗶𝗻𝗶𝘁𝗶𝗮𝗹 𝘁𝗿𝗶𝗮𝗴𝗲, higher autonomy makes sense. L4, and cautiously L5, are reasonable because scale beats precision here. Start at L4. Let agents do the heavy lifting, keep humans reviewing outcomes. ▪️ For 𝗶𝗻𝗰𝗶𝗱𝗲𝗻𝘁 𝗿𝗲𝘀𝗽𝗼𝗻𝘀𝗲 on the right side of the IR cycle, L1 and L2 already work well. L4 is probably the practical ceiling. Approval-based autonomy fits response actions where speed matters but accountability still does. ▪️For 𝗱𝗲𝗲𝗽 𝗶𝗻𝘃𝗲𝘀𝘁𝗶𝗴𝗮𝘁𝗶𝗼𝗻 𝗮𝗻𝗱 𝗮𝗻𝗮𝗹𝘆𝘀𝗶𝘀, autonomy should drop. This is judgment-heavy work. L2 to L3 feels right. Agents assist and lead parts of the flow, but they should actively consult analysts. ▪️For 𝘁𝗵𝗿𝗲𝗮𝘁 𝗵𝘂𝗻𝘁𝗶𝗻𝗴, between L1 to L3. Hunting is exploratory. Collaboration beats full automation. ▪️For 𝗱𝗲𝘁𝗲𝗰𝘁𝗶𝗼𝗻 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴, mostly L2. Possibly L4 only if you have a very mature, well-governed detection lifecycle. Autonomy is a double-edged sword. The goal isn't always L5; the goal is the right level of control for the specific risk profile of the task. Chasing L5 everywhere is usually a design mistake, not a strategy. Has anyone else started classifying their security agents this way?