AhnLab drew more than 500 security officials from Singapore, Malaysia, Taiwan, India and the Philippines to its booth at GovWare 2026 in Singapore. The company showcased its AhnLab AI Plus brand, including an AI Security Operations Center that uses multiple AI agents to triage alerts, analyze threats and make response decisions. The event ran from the 13th to the 15th, and the AI SOC drew particular interest from attendees.
Key Takeaways
- AhnLab showcased its AI SOC at GovWare 2026 in Singapore, attracting over 500 security officials from five Asia-Pacific countries.
- The AI SOC uses multiple AI agents to handle alerts, analyze threats and support faster response decisions under the AhnLab AI Plus brand.
AhnLab displays AI SOC at Singapore security event
AhnLab showed its AI security solutions at GovWare 2026 in Singapore. The company featured its AI Security Operations Center (SOC) under the brand AhnLab AI PLUS. More than 500 security officials from Singapore, Malaysia, Taiwan, India and the Philippines visited the booth. The AI SOC uses multiple AI agents to handle security alerts and response decisions.
AhnLab showcases AI security to Asia-Pacific buyers
AhnLab took part in GovWare 2026 held in Singapore from the 13th to the 15th. The company highlighted its integrated brand AhnLab AI Plus, including the AI security operations center (SOC). Over 500 security officials from Singapore, Malaysia, Taiwan, India and the Philippines visited the booth. The AI SOC drew particular interest because it uses multiple AI agents to triage alerts and support faster decisions.
AhnLab introduces AI security solutions in Singapore
AhnLab announced on the 18th that it introduced AI-based security solutions at GovWare 2026 in Singapore. The company presented its AI Security Operations Center (SOC), IT-OT converged cyber-physical system (CPS) security and threat intelligence (TI) under the brand AhnLab AI Plus. More than 500 security officials from Singapore, Malaysia, Taiwan, India and the Philippines visited the booth. The AI SOC uses multiple AI agents to select alerts, analyze threats and make response decisions.
AI reshapes microdramas as vertical video grows
AI is changing the microdrama industry by helping with editing and even replacing actors. Vertical video formats are making a generational jump in how short serialized stories are made for smartphones. Director Paddy Ma Wendi has worked on vertical dramas since 2023 and sees how fast the changes are happening. Digital content farms are now producing many short productions using these new methods.
Cost per successful outcome is the AI metric CIOs need
SoftwareOne says CIOs should track cost per successful outcome instead of just tokens or licenses. This metric divides total AI run cost by the number of outputs actually accepted and used. A more capable model with fewer errors can be cheaper per result than a cheaper model with many rejected outputs. The article also calls for budgets per agent, alerts for runaway loops and value thresholds to guide AI spending.
Meta removes controversial AI training provision after backlash
Meta removed a provision from its commercial terms on October 17 that would have allowed it to scrape content from websites to train its AI models. Legal experts including James Grimmelmann from Cornell pointed out the provision could have overridden website blocking rules. Meta clarified that the change only applied to ads and business products and did not alter how it uses web crawlers. Content owners can still use robots.txt to signal their preferences.
World Bank report urges AI-driven disaster preparedness in Sweden
A 2026 World Bank and GFDRR report warns that Sweden lacks a unified system to collect and share disaster data despite having extensive records across agencies and universities. Fragmented information, weak coordination, and funding gaps limit Sweden's ability to prepare for floods, wildfires, landslides, and security threats. The report recommends building a national risk and resilience data system supported by artificial intelligence and stronger governance. Events like Storm Hans and the Stenungsund landslide in 2023 highlighted the need for better preparedness.
New benchmark reveals AI agents struggle with long-term decisions
Researchers created Taste-Bench, a benchmark with 502 questions that tests whether AI agents make good choices during multi-step tasks rather than just completing them. The best models tested picked the better path only about 60 percent of the time on binary decision questions. Accuracy drops to near random guessing when key evidence only appears after extensive later work. However, training models using a teacher-student method improved success rates on real tasks from 14.6 to 33.7 percent.
Broadcom CEO predicts years of AI demand with investment risks
Broadcom CEO Hock Tan predicted that AI compute demand could exceed supply through 2029 or 2030 due to rapid AI adoption across industries. He emphasized that capital constraints and returns on AI investment remain key risks for the industry. Tan noted that growing AI use in healthcare, finance, and manufacturing drives the need for high-performance computing. He stressed the importance of careful planning and strategic investment in the AI sector.
Professor Kit Yates explores how AI is transforming mathematics
Professor Kit Yates examines the role of artificial intelligence in mathematics in a video exploring whether AI could replace human mathematicians or serve as a revolutionary discovery tool. The presentation separates science fiction from science fact regarding AI's potential in the field. It highlights both the possibilities and limitations of using AI for mathematical research and problem solving. The discussion covers how AI tools are changing the way mathematicians approach complex problems.
Fifty Years of The Selfish Gene and AI's Role in Science
Richard Dawkins' book The Selfish Gene turned 50 this year. The book teaches that genes are the main drivers of evolution. Recently, Dawkins discussed how AI helps science. Tools like AlphaFold can now predict protein shapes from DNA. This shows AI can solve big problems in biology and medicine.
NVIDIA Structured Data Models Enable Zero-Training Predictions
NVIDIA released a library called Structured Data Models for making predictions from tables. It includes models like TabICLv2 and KumoTabular that learn from examples without extra training. The guide explains how to install the software and fix a common error with pyg-lib. Users can run the models on CPU to predict data from single tables or whole databases.
Sources
- AhnLab Unveils AI SOC at Singapore Security Event, Accelerating Asia-Pacific Push
- AhnLab showcases AI security at Singapore GovWare, courts Asia-Pacific buyers
- AhnLab announced on the 18th that it introduced artificial intelligence (AI)-based security solution..
- How AI is reshaping microdramas as video format takes a ‘generational jump’
- Cost per successful outcome: the AI number Singapore CIOs should be reporting
- Meta deletes provision after experts warn against scraping, AI training
- Can Artificial Intelligence Help Sweden Predict Disasters and Protect Its Critical Services?
- AI agents have bad taste in long-term decisions and a new benchmark proves it
- Broadcom's CEO Sees Years of AI Demand But Warns of a Bigger Threat to the Boom
- How Artificial Intelligence is Revolutionizing Math
- Fifty Years After The Selfish Gene, AI Tests Humanity’s Future
- How to Make Zero-Training Tabular Predictions with NVIDIA Structured Data Models
Comments
Please log in to post a comment.