Tesla's robotaxis are reportedly remotely driven by humans, sometimes
In a letter shared with Senator Ed Markey (D-Mass.), Tesla admitted that its robotaxis are sometimes driven remotely by human operators, Wired reports . Competing self-driving car companies sometimes rely on human operators to tell robotaxi software how to get itself unstuck, but letting operators actually drive those cars remotely is more unusual. "As a redundancy measure in rare cases … [remote assistance operators] are authorized to temporarily assume direct vehicle control as the final escalation maneuver after all other available intervention actions have been exhausted,” Karen Steakley, Tesla’s director of public policy and business development, shared in a letter to Markey. In those situations, operators are reportedly able to take over Tesla's robotaxis when they're moving at spe
In a letter shared with Senator Ed Markey (D-Mass.), Tesla admitted that its robotaxis are sometimes driven remotely by human operators, Wired reports. Competing self-driving car companies sometimes rely on human operators to tell robotaxi software how to get itself unstuck, but letting operators actually drive those cars remotely is more unusual.
"As a redundancy measure in rare cases … [remote assistance operators] are authorized to temporarily assume direct vehicle control as the final escalation maneuver after all other available intervention actions have been exhausted,” Karen Steakley, Tesla’s director of public policy and business development, shared in a letter to Markey. In those situations, operators are reportedly able to take over Tesla's robotaxis when they're moving at speeds around 2mph or less, and then drive the car at up to 10mph if software permits it.
Engadget has contacted Tesla to confirm the details shared in Steakley's letter. We'll update the article if we hear back.
As Wired notes, that's a bit different than how other self-driving car companies handle human intervention. For example, Waymo's Driver software can call on human help — Waymo calls them "fleet response" — to offer context and answer questions to help it navigate complicated driving situations. The company claims these workers never drive the robotaxi themselves, but they are able to see the car's environment through its sensors to help it get unstuck. Self-driving car companies typically avoid remote operation, Wired writes, because technical limitations like latency and the limited perspective of a robotaxi’s sensors can make it hard to drive them easily and safely.
Tesla's approach to self-driving has always cut against the grain, though. Whereas competitors continue to rely on a mix of radar and other sensors to navigate, Tesla has exclusively focused on using cameras for its Full Self Driving (FSD) system. The company has also had to deal with a number of high-profile crashes related to FSD, which prompted a probe by the US National Highway Traffic Safety Administration in October 2025.
The company launched its robotaxi service in Austin, Texas in June 2025, in a limited capacity and with human safety drivers sitting in the driver's seat in case of emergency. Tesla is also reportedly testing rides without safety drivers in the same area, which might be why it has contingencies for remote operators to step in.
Sign in to highlight and annotate this article

Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
launchavailableupdateLLM Cost Tracking and Spend Management for Engineering Teams
<p>Your team ships a feature using GPT-4, it works great in staging, and then production traffic hits. Suddenly you are burning through API credits faster than anyone expected. Multiply that across three providers, five teams, and a few hundred thousand requests per day. Good luck figuring out where the money went.</p> <p>We built <a href="https://git.new/bifrost" rel="noopener noreferrer">Bifrost</a>, an open-source LLM gateway in Go, and cost tracking was one of the first problems we had to solve properly. This post covers what we learned, how we designed spend management into the gateway layer, and what the alternatives look like. You can get started with the <a href="https://docs.getbifrost.ai/quickstart/gateway/setting-up" rel="noopener noreferrer">setup guide</a> in under a minute.</
The Role of AI in Today's Business Landscape
<p>The Role of AI in Today's Business Landscape</p> <p>In the rapidly evolving landscape of technology, <strong>AI-driven solutions</strong> have emerged as a cornerstone for businesses aiming to enhance their operations and customer engagement. From automating mundane tasks to providing deep insights into consumer behavior, AI is transforming industries and reshaping the future of eCommerce.</p> <p>Understanding AI-Driven Solutions</p> <p>AI-driven solutions involve the use of artificial intelligence technologies to improve business functions. These solutions can analyze vast amounts of data, identify patterns, and generate actionable insights that drive strategic decision-making. Companies leveraging AI can boost their efficiency and gain a competitive edge in their respective markets.</
Top 5 Enterprise AI Gateways to Reduce LLM Cost and Latency
<h2> TL;DR </h2> <p>If you're running LLM workloads in production, you already know that cost and latency eat into your margins fast. An AI gateway sits between your app and the LLM providers, giving you caching, routing, failover, and budget controls in one layer. This post breaks down five enterprise AI gateways, what each one does well for cost and latency, and where they fall short. <a href="https://git.new/bifrost" rel="noopener noreferrer">Bifrost</a> comes out ahead on raw latency (less than 15 microseconds overhead per request), but each tool has its own strengths depending on your stack.</p> <h2> Why Cost AND Latency Matter Together </h2> <p>If you're building with LLMs, you have probably already noticed that optimizing for cost alone can tank your latency, and vice versa. Switchi
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
More in Releases
Sweden’s Redpine launches with €1.1 million backing to reduce AI hallucinations through licensed data - EU-Startups
<a href="https://news.google.com/rss/articles/CBMi0wFBVV95cUxOcHdKbWh4N042Z19NVThCNUdVdGFPWkY4ZE56V1NyeE1WR2d6dGY2cG1QU1RUc0syUUJ6bUIzUEktQ1pKSHZJdEo3QUJTLS1IVFk1Rkx5bml6eVZqc2V4QnpTelVPN2ppX05ldllwS2VsN2dtZVlYMEZUVm1feDRtMkQyb1d2NEFHWW9nUFBfS3Y5WEE3UG44SVNFV3hvNE9xUzQyclhoaV9jeFk3UnNoRmp1Tm9IckduRmZMdmZQWDJuVGRGQTdWOVpsZTBTa2FiREgw?oc=5" target="_blank">Sweden’s Redpine launches with €1.1 million backing to reduce AI hallucinations through licensed data</a> <font color="#6f6f6f">EU-Startups</font>
Herd MSL and Publicis Groupe ANZ launch AI search visibility service - Mi-3.com.au
<a href="https://news.google.com/rss/articles/CBMipAFBVV95cUxQdUhMSzZGTUFxdUFHZ1E1TTl3QWdSQmxNX2U1ejZTWndzcEhKendqb3hvUFlDLUUzMEkxQW5yamFHR1M5R3VlbnlueGdreVAwLXBNM25zc2lFRnpEZFJvcEF1bVZKdnI4N3hvd19TdEZDeHFfQTJIbjZqS094c2xSY3ZOMXB1S1hXQ0xLYlI0MjhEa2M4TTZXMVBvQTVwV0dRdmRGMA?oc=5" target="_blank">Herd MSL and Publicis Groupe ANZ launch AI search visibility service</a> <font color="#6f6f6f">Mi-3.com.au</font>
AI Search Barely Cites Syndicated News Or Press Releases - Search Engine Journal
<a href="https://news.google.com/rss/articles/CBMiogFBVV95cUxObGg0U2ZLMDNLTy0yZ1hsYmFIYW9QVnBvcThGdWVRNWVvVUtqS0oyUW5SWWFvVmN5ZDVtOEdYQXVvdmdKbnduQkhJVy1TLXQ3U1ltVThrOWVESDhfeE5YeFlIVWw0TVNWZXBGWXVETGJObmNTZ1pOeXJGYU9hRzNsel9qdEV2cmJjSV93LWJCZTdRV3VjdVh4cE1kanhDcDVwX0E?oc=5" target="_blank">AI Search Barely Cites Syndicated News Or Press Releases</a> <font color="#6f6f6f">Search Engine Journal</font>
White House Releases National AI Legislative Framework, as Competing Congressional Proposals Sharpen the Federal-State Divide — AI: The Washington Report - mintz.com
<a href="https://news.google.com/rss/articles/CBMivAFBVV95cUxPNEdTTWVFS004SGxOLXZYY0Z4TmRfWkpJMGMyVnlQcWR0cW1HSHRIYmN1U1czWjMzMmlyM1ZsSjcxU19lLV93bnhraXRRVGp5Z251TWtTSFFHNmFYdVRhenF1MXNpQmN1TnpJV2E1ZVhTRXJfb1pWWjZBd2tXWWpyaDA5a1p3YWEzdTlMWjdTM2xnX18tVDhGcTFWM19zYm5Pd19UdlVmSXkxRGc2R0c4TTNwYVZiTFJRNlZ3Uw?oc=5" target="_blank">White House Releases National AI Legislative Framework, as Competing Congressional Proposals Sharpen the Federal-State Divide — AI: The Washington Report</a> <font color="#6f6f6f">mintz.com</font>
Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!