SendTech Times
العربية
AINews|July 17, 2026, 03:22 PM
AI SHIFT:

South Korea Targets Year-End Launch For Security AI Model

BySendTech Times AI & Enterprise DeskNewsroom-edited, source-reviewed coverage|Source: The Register
Newsroom brief

The Register reported that South Korea is developing a security-focused AI model for sovereign bug-finding capability, with Deputy Prime Minister Bae Kyung-hoon linking the work to Anthropic Mythos. The report did not disclose the training dataset, project budget, benchmarks or deployment terms.

Verified against source materialEdited by SendTech Times AI & Enterprise Desk
South Korea Targets Year-End Launch For Security AI Model
Image source: The Register

A sovereign bug-finding model is now part of South Korea's AI plan, with the government aiming for a security-capable system online by the end of the year.

Deputy Prime Minister and Minister of Science and ICT Bae Kyung-hoon presented the effort during a policy briefing led by President Lee Jae Myung.

South Korea Wants A Mythos-Class Security Model

Bae described the project as a way for the country to develop its own model for vulnerability discovery.

The target is a system able to rival Anthropic's Mythos, a model associated with finding software bugs.

The government-led approach adds security-related information to the corpus for a locally developed frontier model.

The security-capable version is expected to debut by the end of the year.

US Access Blocks Shape The Sovereign AI Context

Mythos has already been subject to two US access restrictions.

One required Anthropic to make the model available only to American citizens, which the company could not satisfy, and another order kept federal users from accessing it.

Washington later allowed access again.

The earlier blocks remain the source record for Seoul's concern that critical AI systems for security work could be withheld from foreign governments or agencies.

Another Mythos-like effort is being pursued by private US technology companies.

South Korea's project differs because it is tied to a national model programme and a ministerial policy briefing.

Public AI Services Sit Beside The Security Plan

The security model is not the only AI programme moving through Seoul's policy process.

South Korea has sought bids for a chatbot that would be freely available to residents.

Bae's briefing also covered government uses for AI in real-time fake-news detection and public services.

The presidential session listed those topics separately from the security-model plan.

Details regarding the data used for training, funding level, benchmark evidence, access rules, agency rollout dates, and any plan to share the model beyond government use remain undisclosed.

Share this article
inXf

Related articles

More
1Password Gives Claude Session-Scoped Logins Without Model Access
AI

1Password Gives Claude Session-Scoped Logins Without Model Access

SiliconANGLE wrote that 1Password launched a Claude browser integration that lets Anthropic’s agent use approved logins without exposing passwords or one-time codes to the model. The release starts on Mac and leaves payment-card support, personal-detail autofill and independent security validation outside the launch record.

Dream Raises $260 Million For The Hard Sell In Sovereign AI
AI

Dream Raises $260 Million For The Hard Sell In Sovereign AI

Dream has raised $260 million at a $3 billion valuation to expand sovereign AI and national cyber defence platforms across several regions. For Gulf buyers, the missing proof is named deployments, hosting arrangements and procurement scope.

Hitachi Tests Claude for Critical Infrastructure AI
AI

Hitachi Tests Claude for Critical Infrastructure AI

Hitachi partnered with Anthropic to strengthen Lumada 3.0 and bring Claude into mission-critical infrastructure settings. The plan covers HMAX solutions, cybersecurity work, internal deployment to about 290,000 employees and training for around 100,000 AI professionals. The main test is whether safety-focused generative AI can become reliable enough for regulated operational workflows.

OpenAI Keeps GPT-Red Attack Model Private After Prompt-Injection Tests
AI

OpenAI Keeps GPT-Red Attack Model Private After Prompt-Injection Tests

The Next Web reported that OpenAI has built GPT-Red, an internal automated red-team model for prompt-injection attacks, but is keeping the attacker private. The report cited attack success rates above 90% against an older GPT-5 and below 23% against GPT-5.6, while noting that human testers still catch cases GPT-Red misses.

Fujitsu pairs OpenAI and Anthropic deals for enterprise AI push
AI

Fujitsu pairs OpenAI and Anthropic deals for enterprise AI push

Fujitsu announced cooperation with OpenAI and a strategic partnership with Anthropic on the same day. The company plans to use ChatGPT Enterprise, Codex and Claude internally and in customer-facing enterprise AI services. Fujitsu says it will combine external models with Fujitsu Kozuchi, Takane, cybersecurity work and a Forward Deployed Engineer model.

Japan’s Financial Sector Puts Claude Into A Multi-Bank Enterprise AI Test
AI

Japan’s Financial Sector Puts Claude Into A Multi-Bank Enterprise AI Test

Anthropic, NEC and eight Japanese financial companies are moving Claude into a co-creation program focused on financial-service quality, office productivity, cybersecurity and IT modernization.

Keep Reading

More Stories

Latest
Abu Dhabi Tabadul Hub Adds Botswana As First African MarketCapital & PolicyJul 17, 2026Abu Dhabi Tabadul Hub Adds Botswana As First African MarketEconomy Middle East reported that the Botswana Stock Exchange signed an agreement with Abu Dhabi Securities Exchange Group to become Tabadul’s 11th market and first African participant. ADX said the hub covers more than $1 trillion in combined market capitalisation, but the report did not give an integration date or list the first tradable Botswana instruments.Walden Robotics Raises $300M For Wheeled Factory RobotsAIJul 17, 2026Walden Robotics Raises $300M For Wheeled Factory RobotsAI Business reported that Walden Robotics emerged from stealth with $300 million in seed funding, a $1.1 billion valuation and robots already working at a Toyota plant in North America. The company has not named order volumes, contract values, robot counts or deployment dates beyond the Toyota site.Amazon Data Centre In Telangana Draws Rs 1 Lakh Crore Investment RequestCloud & Data CentersJul 17, 2026Amazon Data Centre In Telangana Draws Rs 1 Lakh Crore Investment RequestPress Trust of India reported that Telangana Chief Minister A Revanth Reddy laid the foundation stone for an Amazon data centre at Bharat Future City and asked the company to invest Rs 1 lakh crore by 2034. The report lists Amazon India cloud targets, but not the site capacity, anchor customers or power arrangements.General Compute Lands $400 Million Loan For Inference ChipsCloud & Data CentersJul 17, 2026General Compute Lands $400 Million Loan For Inference ChipsTechCrunch reported that General Compute landed a $400 million Upper90 loan backed by inference-specific chips, extending chip-backed AI infrastructure financing beyond GPUs. The company did not disclose chip quantities, loan pricing, customer contracts, utilisation commitments or revenue.Hugging Face Says AI Agent Drove Production Infrastructure IntrusionCybersecurityJul 17, 2026Hugging Face Says AI Agent Drove Production Infrastructure IntrusionHugging Face said an autonomous AI agent system drove an intrusion into part of its production infrastructure, reaching internal datasets and service credentials. The company said public models, datasets and Spaces were not tampered with, while its assessment of partner or customer data remains unfinished.Nvidia Japan AI Factory Plan Lists 140MW CapacityCloud & Data CentersJul 17, 2026Nvidia Japan AI Factory Plan Lists 140MW CapacityTech Wire Asia reported that Nvidia’s Japan expansion includes a 140-megawatt AI factory using 13,750 Vera CPUs and 27,500 Rubin GPUs for the METI-backed FRONTia Project. The report did not name the site, capital cost, power supplier, construction timetable or first enterprise workloads.Open-Weight AI Model Backdoor Test Costs Less Than $100AIJul 17, 2026Open-Weight AI Model Backdoor Test Costs Less Than $100The Register reported that Katie Paxton-Fear installed a backdoor in an open-weight AI model in about an hour for less than $100. The experiment points to model-poisoning risk, but the cited public examples do not identify a widely deployed poisoned model or affected customers.7-Eleven Philippines Extends Cashless Payments Past 4,000 StoresFintech & Digital PaymentsJul 17, 20267-Eleven Philippines Extends Cashless Payments Past 4,000 StoresCompany records put cashless payment acceptance at more than 1,000 7-Eleven stores at the end of 2025 and more than 4,000 by the end of May 2026.Xi Backs Shanghai AI Governance Body After 29-Country DealAIJul 17, 2026Xi Backs Shanghai AI Governance Body After 29-Country DealCNBC reported that Xi Jinping used a Shanghai AI speech to position China as a Global South technology partner after 29 countries signed an agreement to establish the World Artificial Intelligence Cooperation Organization. The address named cooperation with ASEAN, the African Union and BRICS, but did not identify WAICO members, funding or enforcement powers.US Lawmakers Seek Ban On Chinese Memory PurchasesChips & SemiconductorsJul 17, 2026US Lawmakers Seek Ban On Chinese Memory PurchasesRepresentatives John Moolenaar and George Whitesides urged the Commerce Department to restrict US purchases from YMTC and CXMT as memory prices are expected to stay high through at least 2028.Fireworks Raises $1.5B As AI Platform Reaches $17.5B ValuationAIJul 17, 2026Fireworks Raises $1.5B As AI Platform Reaches $17.5B ValuationFireworks AI raised a $1.5 billion Series D at a $17.5 billion valuation, SiliconANGLE reported. The company says it processes more than 40 trillion tokens a day, while contract values, total GPU capacity and customer commitments tied to the new capital remain outside the public record.Nebius Moves AI Cloud Expansion To Partner-Owned Data CentresCloud & Data CentersJul 17, 2026Nebius Moves AI Cloud Expansion To Partner-Owned Data CentresAI Business reported that Nebius plans to run its AI cloud platform in partner-owned data centres, with revenue shares, licensing fees and committed-capacity deals under discussion.