Industrial Next Retools AI Factory Robots in Minutes as BMW Tests Technology
Industrial Next says its AI robotic stations can retask factory lines in minutes, while BMW Group evaluates its dexterous manufacturing technology in 2026.
Summary
San Francisco-based Industrial Next said on October 13, 2026, that its self-contained robotic stations can retool production lines in minutes rather than weeks, without traditional programming. Proprietary foundation models connect vision, task context and motion, while high-fidelity simulation and demonstrations teach precise, dexterous tasks despite robotics’ limited real-world training data. Stations can be reconfigured, combined and redeployed across products. BMW Group Technology Office will evaluate the technology for manufacturing.
Lukas Pankau and Allen Pan founded Industrial Next in 2021 after working at Tesla, where Pankau was lead electrical architect and Pan led autonomous factory efforts. Pankau later joined Waymo, while Pan became a Neolix chief architect and also worked at Momenta. Industrial Next says its stations already help manufacture an unnamed, top-selling smartphone by threading delicate flex cables, attaching tiny connector caps and peeling protective film from camera modules and buttons. The film-peeling system operates at or above human speed with 100% accuracy, including after transfer to a previously unseen phone model and production line. The company now aims to progress from phone assembly to end-to-end car production without human handling.
Positives
- Production lines can switch tasks in minutes rather than the weeks or months required for conventional programming and physical retooling.
- 100% accuracy at or above human speed was achieved while peeling protective film from camera modules and buttons.
- BMW Group Technology Office will evaluate Industrial Next’s AI-powered dexterous manipulation for manufacturing.
- High-fidelity simulation and demonstrations teach precision tasks without traditional hard-coded robot scripts.
- Industrial Next stations already perform smartphone assembly tasks that manufacturers had failed to automate for years.
Risks & concerns
- Robotics lacks the vast training datasets available to internet-based AI models, complicating flexible physical automation.
- Small product or process changes can make conventional industrial robots fail, causing costly downtime and forcing simpler product designs.
- BMW Group’s work remains an evaluation, with deployment scale and timing still unspecified.
- Industrial Next cannot identify the smartphone customer using its workstations.
- End-to-end car production without human handling remains a long-term goal beyond the company’s disclosed smartphone tasks.