Mobileye Global is a cash-generating driver-assistance chip franchise that is now spending that cash on a founder-led pivot into robotaxis and humanoid robots, just as the founder has told the board he is leaving the chief-executive seat. The second-quarter print was essentially unchanged at the top line, even as shipment volume rose 3 percent and the company again outgrew the production of its largest automaker customers. What the market is pricing is not a broken core. It is skepticism that the next act converts a profitable EyeQ franchise into higher-content systems without burning the cash engine that still funds everything else.
The tension sits in mix and capital allocation rather than demand. Average system price slipped as China export volumes, which carry thinner EyeQ pricing, ran hotter than planned. SuperVision hardware diluted reported gross margin even though unit profit on that product is richer than base driver-assistance chips. Adjusted operating income jumped to $155 million, but the jump is mostly an Israeli research-incentive grant recognized for the entire first half in a single quarter. Strip the catch-up and the first half still expanded the margin against a decline in top-customer production, which is the cleaner read on the chip business.
The next year resolves whether Surround programs, a Stellantis cloud-enhanced win, and a planned own-fleet robotaxi service change the mix, or whether Intel dual-class control and another secondary keep the multiple pinned to a low-growth auto-chip tape. Cash from operations covered the first half. The cash that left the balance sheet left for Mentee Robotics. Does the cash engine fund a real mix shift, or a science project?