Aiv
3 known investors
AIV builds an AI vision-based robotics automation platform aimed at implementing industrial "Physical AI," connecting the platform with robots for setup, operation, and performance management in production settings. It focuses on perception-driven robot automation and control for high-variance tasks, integrating inspection, decision-making, and motion for manufacturing.
Also known as AIVEX · 아이벡스
Investors · 3
Company profile
researched Aug 2026AIVEX (아이벡스), operating at aiv.ai, positions itself as an industrial "Physical AI" company whose stated aim is for industrial sites to see clearly, judge intelligently and act autonomously. Its offering combines industrial AI platform products with AI solutions intended to create a self-improving "Learning Factory" that requires minimal human intervention, unifying what it describes as a fragmented AI Transformation (AX) process into a single operating pipeline.
The platform consists of four connected products. AIVision runs deployed models on the line to judge and classify product condition; AIVData accumulates and manages process images and equipment data in a central hub; AIVOps handles the full model lifecycle from training through deployment using curated data; and AIVot links vision inspection results to robot motion and to control of existing production equipment. The company frames manufacturing AI as an operations problem as much as a model-accuracy problem, citing field issues such as performance degradation when the environment changes, scarce defect data, ambiguous judgements and heterogeneous equipment constraints.
Business model
AIVEX sells industrial AI platform products (AIVision, AIVData, AIVOps, AIVot) together with AI solution and integration work for manufacturing customers, including proof-of-concept engagements and end-to-end system design that spans optics, hardware such as grippers, models and robot control [0][2].
Traction
Reported field results include a 3 ppm (0.0003%) miss rate in bullet exterior inspection for a defense customer, detection of 7-pixel-scale defects in 300M-pixel high-resolution images while high-speed scanning ten faces of large parts over 300mm on a side, depalletizing automation for a large logistics customer after prior vendors could not recognize box heights, and stable line operation for an automotive parts customer that had previously deployed another vendor's AI vision equipment [0][2].
▸Full profile — market position, technology, go-to-market, geography
Market position
AIVEX describes itself as a Global Physical AI Leader and cites deployments where customers replaced or returned from competing AI vision equipment; in a tire application a customer states no AI equipment existed that could inspect the full tire including the inner surface [0][2].
Stated differentiators are data efficiency (a validation case compressing 155,282 training images to 4,284, a 36x reduction), detection performance reported 27.2 percentage points above a competing model, 100% detection of safety-critical defects using out-of-distribution methods for unseen rare defects, and management of over 7,000 production items under a single model structure rather than per-product models. The company also emphasizes closing the loop from inspection to robot motion and existing equipment control rather than stopping at on-screen inference [0][2].
Technology
The company integrates a technology stack across Sensing, Model, Mechatronics, Motion Control and Ops. Sensing covers optical modules for detecting fine defects against noisy backgrounds at production speed and for supporting robot motion that imitates human actions. Model covers generative AI to synthesize scarce defect data, anomaly-detection models trained only on good samples, and Vision and Vision-Language-Action (VLA) models for transferring fine manual tasks to robots; it applies both supervised and unsupervised learning and uses out-of-distribution techniques to catch untrained rare defects. Mechatronics covers inspection hardware and interoperability with robot makers and peripheral devices. Motion Control includes Auto Teaching for handling many product variants plus precision hardware control that links inference results directly to robot actions such as precision assembly and logistics handling. Ops provides automatic retraining and validated model redeployment as lighting or materials change, and automatic collection of judgement images and equipment data into a central hub [0][2].
Go-to-market
The website offers an adoption diagnostic and inquiry path for prospective customers, and describes engagements that begin with proofs of concept before moving to line deployment; references span global large enterprises to specialist manufacturers [0][2].
Manufacturers and logistics operators, including semiconductor back-end and OSAT firms, display makers, battery cell and battery pack producers and a global battery manufacturer, automotive parts makers, a tire manufacturer, a defense-sector customer and large logistics companies [0][2].
Geography
Korean-language operations with a Korean brand name (아이벡스) and customers described as ranging from global large enterprises to specialist manufacturers [0][2].
Compiled by commissioned research from 8 cited public sources — announcements, filings, and press listed under research sources below.
Key figures
latest reportedCompany-reported or press-reported figures, each dated to when it was claimed — not independently audited.
▸Research sources · 8
primary sources listed
- Aivaiv.ai · web
8 public sources were cited for this profile; the first-party ones are listed here.
Frequently asked questions
- What does Aiv do?
- AIVEX is a Korean industrial AI company building vision inspection, data, MLOps and robot-control products for factories.
- Who are Aiv's investors?
- Aiv's investors include Kiwoom Investment, LB Investment, We Ventures.