Moves by the security industry are accelerating toward commercial security services that use large language models (LLMs) to help public and private organisations detect software security vulnerabilities.

IBM, working with OpenAI, unveiled an enterprise artificial intelligence security service. The service finds code vulnerabilities, identifies paths for real-world exploitation and helps users respond before security issues occur. Japan's SoftBank Group launched a new cybersecurity product based on OpenAI models, called Patching as a Service. SoftBank Group Chairman and CEO Masayoshi Son (손정의) said at a presentation for corporate customers in Tokyo, "We will build a system that can protect Japan's core infrastructure. We will use a new weapon called OpenAI for defence."

ㆍFind it before hackers… IBM and OpenAI unveil an 'AI security service' that tracks code vulnerabilities ㆍSoftBank launches a new cybersecurity product based on OpenAI models

OpenAI updated its cybersecurity model GPT-5.5-Cyber and expanded the scope for verified security firms to use it. OpenAI is also strengthening the OpenAI Daybreak Cyber Partner Programme, which supports verified cybersecurity companies in using its model to secure customer environments. The programme centres on enabling participating security firms to use GPT-5.5 Trusted Access for Cyber in the products and services they provide to customers.

ㆍOpenAI updates GPT-5.5-Cyber... expands cooperation with security firms

As cyber attackers actively use AI, phishing crime is increasingly becoming automated. An analysis said phishing attacks that bypass traditional identity verification tools have surged by nearly 15 times this year, and this is not unrelated. Five Eyes, a signals intelligence cooperation chain among intelligence agencies in Australia, the United States, Britain, New Zealand and Canada, warned that destructive AI cyber attacks targeting governments and businesses could become reality within months.

ㆍCyber attacks enter the era of AI automation... subscription-based crime platforms spread ㆍ"Destructive AI cyber attacks could become reality within months"... Five Eyes warns

It also summarised moves and issues surrounding security by companies at home and abroad.

China's cybersecurity company 360 Security Technology developed its own AI security tool to counter Anthropic Mythos. 360 Security unveiled security AI systems like Mythos and AI that automates cyber defence and incident response. F5 is accelerating its integrated security platform strategy targeting AI environments. F5 also acquired SurePassAI, a platform company that detects the use of AI tools and platforms and enforces cybersecurity policies.

ㆍChina's 360 Security says it developed a rival to Mythos… unveils vulnerability-detection AI ㆍConditions for an integrated AI security platform... "block three control points" ㆍF5 acquires SurePassAI... secures visibility into AI traffic

Cisco agreed to acquire WideField Security, an identity lifecycle security company. The move aims to strengthen Splunk agentic SOC functions. Barracuda Networks launched integrated email security functions that use AI to automatically detect and remove email threats. Cybersecurity startup Ent raised $100 million in seed funding. Ent developed an "intent-aware workspace security" platform that analyses the behaviour of employees, AI agents and applications. Israeli cybersecurity startup Dream raised $260 million in funding. Dream specialises in national cyber defence platforms. Its Sphere platform combines AI detection, cyber intelligence and digital twin technology to help detect and respond to nation-level cyber threats.

ㆍCisco to acquire identity security company WideField... strengthens Splunk agentic SOC ㆍBarracuda Networks launches AI-based email security product... blocks malicious emails in real time ㆍEnt raises $100 million... develops intent-based AI agent security platform ㆍDream raises $260 million... expands national cyber defence platform

ITCEN Group's security unit ITCEN PNS will take part in the standardisation process at home and abroad for the domestically developed post-quantum cryptography (PQC) algorithm NTRU+. AI services and solutions company Bespin Global signed a partnership with Wiz, a global cloud security firm under Google. ITCEN Global signed a strategic memorandum of understanding with BTQ Technologies, a Nasdaq-listed quantum technology company, to build PQC and quantum-safe security infrastructure. Application security company Sparrow launched an AI-generated code security assistant called Sparrow MCP to help AI models generate safe code. KT unveiled E2E Quantum Security, a future network security concept to respond to changes in AI and quantum technology environments.

ㆍITCEN PNS to support standardisation of homegrown post-quantum cryptography 'NTRU+' ㆍBespin Global cooperates with cloud security firm Wiz ㆍITCEN Global partners with global quantum technology company BTQ on quantum security business ㆍSparrow launches MCP-based AI-generated code security assistant ㆍKT unveils E2E Quantum Security network concept to respond to quantum computing