The patient in Jinjiang had lost two front teeth to an accident, and the bone behind them had gone with them. At a county-level hospital in Fujian, the usual answer would have been a long wait: pull what remained, let the socket heal, graft bone, wait months, then drill.
Instead the surgeon spent the morning planning on a screen — implant position, angle, depth — and then let a robotic arm hold the drill. The patient left with temporary teeth the same day.
The system was a dental implant robot (口腔种植机器人), a category that has moved in five years from a laboratory curiosity to a line item in Chinese hospital budgets, and is now being exported.
Key takeaways
- Beijing Yakebot Technology (雅客智慧) says its autonomous dental implant robot has been used in more than 30,000 implant placements worldwide, with 120-plus units sold and 20,000-plus surgeries completed as of late August 2026.
- The system targets 0.3–0.5 mm placement accuracy, with real-time optical tracking and force feedback that can stop the arm when resistance looks wrong.
- On 29 July 2026, Singapore's Foundation Healthcare opened a centre at Farrer Park Medical Centre using the robot — the country's first — with a research and training tie-up with the National University of Singapore's dental faculty.
- In China the robot has spread beyond top-tier hospitals to county-level sites; Yakebot says about half its sales now go to private dental clinics.
What the robot actually does
The device is a supervised machine tool, not an autonomous surgeon. A CBCT scan and an intraoral scan are fused into a three-dimensional plan; the surgeon approves the plan; the arm then drills along that path while cameras track the patient's jaw and load cells watch for unexpected resistance. The dentist keeps a foot pedal and a stop button, and makes every clinical decision.
That division of labour is why dentistry became one of the first surgical fields to go robotic. Bone is a fixture, an implant is a screw, and the task is essentially high-precision drilling in a cramped, wet, poorly lit space where the surgeon's view is often blocked by the patient's own lip.
At Sun Yat-sen University's dental hospital in Guangzhou, surgeons recently used a domestic autonomous system for a zygomatic implant — anchoring a long implant in the cheekbone because the upper jaw had resorbed too far to hold one. The drill path passes close to the eye socket and the sinus, and the angle is steep. That is exactly the geometry where a human hand and a mirror are weakest and a tracked arm is strongest.
From county hospitals to Farrer Park
Yakebot's own figures trace a familiar Chinese diffusion curve. It was approved through the National Medical Products Administration's innovative-device pathway, won European CE certification in 2025, and now reports installations in Vietnam, Hong Kong and Singapore.
The Singapore entry is the more consequential one commercially. Foundation Healthcare, a Singapore private-hospital group, is pairing the robot with an academic collaboration rather than simply installing a machine — the pattern that tends to precede reimbursement and training standards.
Domestically, the interesting number is not the top-tier hospitals but the county-level ones. Jinjiang in Fujian, a county-level city, performed its first robot-assisted implant in August. A technology that begins in teaching hospitals and ends in county clinics has usually crossed a cost threshold.
Why precision is the easy part
Implant dentistry has a complication rate problem that is hard to see from the outside: a millimetre of error near the mandibular nerve means numbness, and a few degrees of tilt near the sinus means a graft that fails quietly two years later. Robots do not eliminate those errors, but they make them repeatable and auditable — every case leaves a plan and a log.
That has a second-order effect that matters more than accuracy claims. Once a procedure is planned digitally and executed mechanically, it can be taught, delegated and quality-checked in ways that freehand surgery cannot. It is the same logic that turned cataract surgery into a high-volume, low-variance business.
Honest limitations
The 30,000-implant figure, the 120-unit installed base and the 20,000-surgery count are company-reported and have not been independently audited; no randomized trial comparing robotic and freehand implant survival is cited here. The 0.3–0.5 mm accuracy figure is a technical specification, typically measured under controlled conditions, not a clinical outcome rate. "Autonomous" in this context means the arm executes an approved plan under supervision — the surgeon is not absent. Cost per case, reimbursement and long-term failure rates are undisclosed, and Singapore's private-pay setting says nothing about access in public systems. A crowded field of domestic oral surgical robots has now been approved in China, so claims of firsts are marketing as much as history.
What readers can do now
- If you are considering an implant in a complex site — front tooth, nerve proximity, sinus lift — ask whether the clinic plans in 3D and what its revision rate is, robot or not.
- If you run a dental group, compare the robot on workflow, not accuracy: planning time, chair time and lab turnaround decide payback.
- If you follow medical robotics, watch county-level adoption. It is a better signal of cost curves than any conference demo.