Elon Musk now spends nearly 50% of his time on Tesla earnings calls talking about artificial intelligence, robotaxis, and Full Self-Driving, according to a seven-year transcript analysis by New York-based financial research firm Hudson Labs. That share is up from 15% to 20% in 2022, and it comes even as Tesla shipped nearly half a million cars last quarter and still generated 70% of its revenue from car sales.
Hudson Labs pulled Tesla earnings-call transcripts back to 2019 from S&P Market Intelligence and ran them through its Co-Analyst tool, tagging each sentence by topic and counting frequency. The dataset covers seven years of calls, tracking how executive attention has migrated from the automotive business toward AI and robotics projects that do not yet generate meaningful revenue.
Musk has been explicit about the reframing. On the Q1 2024 call, he told investors the auto-company framework was the wrong one and that skeptics of Tesla's autonomy roadmap should not own the stock. That message has scaled directly with airtime: as Tesla's car business hit competitive pressure from legacy automakers and Chinese entrants in 2024, Musk's on-call focus tilted further toward AI.
Key facts
- 01Musk now spends nearly 50% of Tesla earnings calls on AI, robotaxis, and Full Self-Driving, up from 15-20% in 2022.
- 02Tesla shipped nearly half a million cars last quarter and still gets 70% of revenue from car sales.
- 03Optimus went from under 2% of Musk's earnings-call remarks in 2022 to nearly a third of his Q3 2025 call.
- 04On the Q3 2025 call, Musk spent less than 20% of his time on the automotive business.
- 05Other Tesla executives spend about 30% of call time on cars, down from nearly 50% or more before 2024.
The Optimus humanoid robot, first revealed in 2021, shows the sharpest swing. In the year after the reveal, Musk spent 2% or less of his call time on the project. Over the past year he has spent at least 10% of his remarks on Optimus, and on the Q3 2025 call it took up nearly a third of his focus. On that same Q3 2025 call, he spent less than 20% of his time on the automotive business.
Across the last three years, robotics has climbed steadily as a topic while cars and manufacturing have receded. Musk now spends less than a third of his earnings-call time on the vehicle business — the segment that still funds the company. The gap between what pays the bills and what dominates the narrative has widened every quarter.
Other Tesla executives have been slower to follow. Chief financial officer Vaibhav Taneja and vice president of engineering Lars Moravy still spend around 30% of their call time on the automotive business on recent calls, with AI, robotaxi, and Full Self-Driving trailing as their next-most-common topics. Before 2024, those executives spent nearly 50% of their time, or more, on making and selling cars.
The lag likely reflects operational reality. Taneja and Moravy own the numbers investors can model — deliveries, margins, factory ramps — while Musk owns the story. When the operating executives do venture into the AI narrative, they match the register.
The stock market has largely gone along with the reframing. Tesla trades at a multiple that only makes sense if a meaningful share of its future value comes from autonomy and robotics rather than vehicle unit economics. The Hudson Labs data quantifies what shareholders already price in: Musk is selling an AI company that happens to build cars, and he is spending his airtime accordingly.
The counterweight is that Optimus and robotaxi have yet to generate disclosed revenue at scale, and Full Self-Driving remains a driver-assist product rather than a fully autonomous system. Competition in humanoid robotics is intensifying from firms including Figure and 1X, and Chinese automakers continue to erode Tesla's EV share in key markets. If AI-driven revenue does not materialize on the timeline the calls imply, the gap between narrative airtime and cash flow becomes a valuation problem rather than a storytelling one.
For the broader AI market, Tesla's earnings-call shift is a data point in a pattern: hardware companies with mature core businesses are increasingly pricing themselves on AI optionality rather than segment performance. Nvidia set the template on the supply side; Tesla is testing whether the same premium holds on the applied side, where autonomy and humanoid robots have to actually ship. The next several quarters of Optimus deliveries and robotaxi mileage will decide whether the 50% of airtime was prescient or premature.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
