South Korea’s next AI unicorns stand out at KPAS 2024
The 20 finalists are startups with promising technologies in the AI cloud, AI model, AI chip and AI service sectors
By Oct 17, 2024 (Gmt+09:00)
LG Chem to sell water filter business to Glenwood PE for $692 million


KT&G eyes overseas M&A after rejecting activist fund's offer


Kyobo Life poised to buy Japan’s SBI Group-owned savings bank


StockX in merger talks with Naver’s online reseller Kream


Meritz backs half of ex-manager’s $210 mn hedge fund



South Korea’s major telecom company KT Corp. and the country’s leading financial newspaper The Korea Economic Daily have teamed up to scout out the country’s next unicorns with promising artificial intelligence technologies expected to innovate the country’s AI ecosystem.
At Korea Promising AI Startups (KPAS) 2024, jointly hosted by KT and The Korean Economic Daily on Thursday, 20 AI startup finalists were unveiled with great fanfare.
Through thorough reviews by venture capitalists and local AI industry experts, they were chosen for their technologies expected to create great synergy with other industries such as cloud and chip technologies and open new markets.
The startup finalists -- lauded to include Korea's next AI unicorns -- have estimated enterprise values ranging between 100 billion won ($73 million) and 700 billion won, and some of them are already actively leading digital transformation at various industrial sites in Korea.
AI SERVICE
Nearly half of the finalists offer AI services designed to empower other industrial processes through automation and prediction systems.

Ineeji offers AI programs that can improve efficiency across manufacturing processes and save energy.
Daim Research has developed an AI and digital twin technology-powered factory platform that enables autonomous logistics robot operation and management.
Datarise offers AI-powered data analytics and customer relationship management (CRM) services for financial and retail customers.
MarqVision, a startup developing an AI-driven intellectual property (IP) protection platform, saw its sales in the first quarter this year grow 30 times from three years ago. It counts international luxury powerhouse LVMH Moët Hennessy Louis Vuitton (LVMH) among its clients.
It recently raised 22 billion won in a Series A round and used some of the proceeds to develop generative AI-backed Marq AI.
Imagoworks offers automated dental treatments using AI, computer-aided design (CAD) and cloud technologies, while Roen Surgical has developed Zamenix, an advanced flexible endoscopic surgical robot system.

Nearthlab develops AI-controlled, autonomous drones, and its inspection and counter-unmanned aircraft systems are lauded for their excellent performance.
Muhayu, known for its flagship AI-powered plagiarism detection service Copy Killer, has recently launched a Japanese version of GPT Killer designed to detect generative AI plagiarism.
AI CLOUD
Five startups that develop, distribute and manage AI models based on cloud computing were also selected.
Lablup offers a program used by AI developers to build, train and deploy machine learning models using graphic processing units (GPUs).
Dalpha offers cloud-backed customized AI services for various business-to-business clients, while Vessl AI operates a MLOps platform.
Allganize develops large language models for different industries, and Selectstar, an all-in-one data service provider, is currently developing an AI-backed mental health care management service as part of the Korean government’s project.

AI MODEL AND AI CHIP DEVELOPERS
Nota AI is also recognized for its smaller AI models that support small devices in the AI model sector.
It was joined by three peers – Superb AI, operator of a platform that labels, cleans, analyzes and manages data needed to build an AI model; Aurumplanet, operator of AI-backed custom search engine Liner; and Friendli AI, offering a service to optimize client’s LLM to reduce inference time and maximize performance.
Hyper Accel is accoladed for its LLM processing unit (LPU), an AI chip designed for transformer-based LLM.
It is joined by two other AI chip-developing peers in the AI chip sector – MetisX, which develops a compute express link (CXL)-based memory chip needed for large-scale data processing such as AI big data, and Dnotitia, a vision language model developer, which is currently co-developing an AI-supported medical imaging analysis system with one of the country’s major university hospitals.
The 20 finalists of KPAS 2024, the first event of its kind, will receive support from KT and The Korean Economic Daily for fundraising, consulting services for technology development and commercialization, as well as marketing and promotion services.
Write to Joo-Wan Kim at kjwan@hankyung.com
Sookyung Seo edited this article.
-
Artificial intelligenceAI startup MarqVision raises $16.5 mn to close Series A round
Oct 04, 2024 (Gmt+09:00)
1 Min read -
Artificial intelligenceLiner ranks 9th in generative AI services selected by a16z
Aug 23, 2024 (Gmt+09:00)
1 Min read -
Korean startupsMLOps startup Vessl AI eyes 50% sales from US in 2025
Aug 05, 2024 (Gmt+09:00)
5 Min read -
Artificial intelligenceAI plagiarism detection startup Muhayu eyes IPO in 2026
Jul 03, 2024 (Gmt+09:00)
1 Min read -
Upcoming IPOsKorea’s vision AI startup Superb AI eyes IPO in H1, 2026
Apr 18, 2024 (Gmt+09:00)
2 Min read -
Korean startupsKorean AI startups Allganize, Neurophet raise over $35 mn
Nov 08, 2023 (Gmt+09:00)
2 Min read -
Artificial intelligenceNaver-invested Nota AI named global leader in AI optimization
Oct 20, 2023 (Gmt+09:00)
1 Min read -
Artificial intelligenceSilicon Valley-based AI startup Allganize revs up to go public in Japan
Aug 17, 2022 (Gmt+09:00)
2 Min read