South Korea picks SK Telecom, KT, and Kakao to run AI-for-All

Share
South Korea picks SK Telecom, KT, and Kakao to run AI-for-All

South Korea has picked SK Telecom, KT, and Kakao to run "AI for All," a public-utility program that will give every Korean citizen free, unlimited access to frontier AI chatbots and agents by the end of 2026 — and the government is putting real money behind it.

South Korea just made AI a public utility. The Ministry of Science and ICT on Friday selected three operator consortiums — led by SK Telecom, KT, and Kakao — out of six applicants to run "AI for All," a program that will hand every Korean free, unlimited access to general-purpose AI chatbots and specialized agents with no usage cap. The government will supply a combined 512 Nvidia B200 GPUs this year, then from 2027 will subsidize part of the nationwide service costs directly out of the budget. The contracts are due to be signed in September, with a beta period before the formal launch by year-end. It is the first major economy to treat consumer AI access the way it treats broadband or electricity — a utility-grade entitlement rather than a paid product.

Naver's absence is the subtext. Korea's largest web platform deliberately sat the project out. The company, which runs the HyperCLOVA X foundation model, was already dropped from the government's separate sovereign foundation model program in January over data-independence concerns, and reportedly concluded that AI for All's tight timeline and the requirement to integrate other vendors' models offered a poor return on the investment. The three winners are filling that vacuum with a more nationalistic stack: SKT's consortium leans on its A.X K2 model (688 billion parameters, built under the sovereign-model program) plus Solar Open 2 from Upstage and K-EXAONE 2.0 from LG AI Research; KT's group runs on KT's Mi:dm K Pro 2.5 plus Upstage's Solar Pro 4; Kakao's bundle uses its own Kanana model alongside LG's EXAONE family. Operators are required to mix their own models with those from other companies — a deliberate move to keep the program from becoming a single-vendor showcase.

The differentiator is access, not technology. Each consortium won on a distinct delivery model rather than model performance. SKT's pitch is a task-execution agent reachable by phone call and text message, not just apps — explicitly aimed at older users who don't live in app stores. KT's group is wiring the AI into Daum's portal and partner platforms like Musinsa and Zigbang, with the chatbot handling the full arc from search to checkout in a single conversation. Kakao is making KakaoTalk the on-ramp and bundling healthcare (via Seoul National University Hospital), finance (Shinhan Bank), and caregiving into one integrated agent surface. If it works, Korea will be the first country where the average citizen's default AI assistant is the one their government subsidized, not the one their phone came with.

What to watch: how fast the three operators can stand up 512 B200s of capacity, and whether the per-user compute bill stays inside the government's subsidy envelope once usage patterns stop looking like a beta.

Should national governments be the ones underwriting free consumer AI? Tell us in the comments.

Read more

Akhetonics says its all-optical CPU reaches a customer in 2026

Akhetonics says its all-optical CPU reaches a customer in 2026

Light-based computing keeps promising more than it delivers — but one Munich startup has just put a date on its bet, and the interview laying it out is doing the rounds on Hacker News this week. Akhetonics says it will deploy its first commercial machine with a major customer by the end of 2026, with several more planned for 2027. The company, founded by Michael Kissner and Leonardo Del Bino, is building a computer where data enters as light, is switched as light, and circulates through memory

The Week in AI — October 5–11, 2026

The Week in AI — October 5–11, 2026

Every big claim this week turned out to rest on fine print more interesting than the headline: revenue only the company reporting it can define, safety tests sandboxed while the product keeps the web, and a Pentagon phase-out nobody would confirm until reporters kept asking. The week's top 5 1. OpenAI's revenue was $20 billion below the numbers everyone quoted — and the gap was definitional. The Financial Times reported Thursday that OpenAI's annualized revenue runs roughly $20 billion unde

Drone strike shuts a third Yandex data center, taking YandexGPT offline

Drone strike shuts a third Yandex data center, taking YandexGPT offline

Russia's largest tech company is learning what the AI era's infrastructure war looks like from the receiving end — three data centers in four days, and with them much of the cloud layer Russian businesses run on. A Ukrainian drone strike knocked out Yandex's data center in Vladimir early Sunday morning, the third of the company's facilities hit since October 8. The site — reported at roughly 50 MW and designed for about 2,880 server racks — stopped operating completely after the attack, Yandex

Agent teams cost up to 5x more, barely score higher

Agent teams cost up to 5x more, barely score higher

The multi-agent hype train hit a benchmark this weekend — and the grid and the trucking regulators had quiet weeks of their own. Vals AI put agent teams head-to-head with single agents on its Vibe Code Bench, and the teams cost between 1.8 and 5.1 times more for almost no extra quality. The evals company ran GPT-6 Sol and Claude Opus 5.5 solo and in teams across 50 apps at two reasoning efforts; out of four comparisons, only one was statistically significant — Sol at medium effort, where the t