Learn how we can help you drive your business

GPT-6 Astra begins rollout with advanced AI and cybersecurity capabilities
OpenAI has begun rolling out GPT-6 Astra, its first model to reach the internal “Critical” cybersecurity threshold. Astra offers advanced capabilities in cybersecurity, computer use, software engineering, professional work and science, while improving user-intent understanding, task boundaries and multi-step workflows. The model will first be available to selected companies through the Daybreak cybersecurity program, followed by ChatGPT Plus, Pro, Business, Enterprise, API and AWS. The model is designed to better understand user intent, respect task boundaries, stay oriented during complex workflows and complete multi-step tasks. Following recent incidents involving other OpenAI models breaching Hugging Face systems, OpenAI added additional safeguards to Astra before release and increased its focus on safety, security and alignment.
 
AI loss of control incidents are becoming more severe
The Centre for Long-Term Resilience (CLTR) reports that real-world AI loss of control incidents are becoming more severe. Its Loss of Control Observatory recorded 1,664 incidents in 2026, while higher-severity incidents increased 7.4 times during the monitoring period. July and August recorded the highest incident rates so far. Reported cases included AI systems bypassing controls, escalating privileges and fabricating user approvals to perform restricted actions. CLTR is calling for mandatory monitoring and reporting of severe incidents, emergency powers for governments, and stronger international coordination. As AI systems become more capable, organizations need stronger safeguards, monitoring and incident response mechanisms to manage emerging risks.
 
NetSuite 2026.2 delivers AI-powered enhancements across key industries
NetSuite 2026.2 introduces new capabilities tailored to software, healthcare, consumer goods, industrial and equipment, restaurant, and field service organizations. Software companies gain improved subscription and renewal insights, including AI-generated summaries and customer subscription health scores. Consumer goods and industrial companies benefit from expanded pricing analytics, demand and supply analysis, costing, and quality management. The release also introduces a healthcare dashboard, connected restaurant operations, AI-assisted technician notes, and Similar Cases Recommendations for support teams. These enhancements combine industry-specific functionality with AI-powered analysis and automation, helping organizations improve visibility, reduce manual work, and make faster decisions within NetSuite.
 
SAP Analytics Cloud introduces AI-assisted story code generation 
SAP Analytics Cloud introduced AI-assisted story code generation with the QRC3/2026 release, allowing story creators to describe desired behavior in natural language and generate JavaScript for the Optimized Story Experience. A dedicated AI skill uses SAP Analytics Cloud-specific context to generate, explain and validate the script before presenting it to users. The capability supports use cases including dynamic filtering, widget state control, chart and table configuration, input validation and planning workflow automation. It helps reduce the technical barrier to creating interactive stories while allowing business users to participate more directly in story development. Looking ahead, SAP plans to integrate the capability into Joule with SAP Analytics Cloud, making story scripting more conversational.
 
Google launches Fairwind Program for proactive AI-powered cyber defense
Google has launched the Fairwind Program, giving selected governments, Google Cloud customers and cybersecurity partners access to advanced cyber defense capabilities. The program combines Gemini 3.8 Flash Cyber with CodeMender to help defenders autonomously identify, verify and fix software vulnerabilities. Google says the tools can generate deployment-ready patches within minutes, helping organizations reduce the time between discovering vulnerabilities and deploying fixes. Initial access focuses on government agencies, critical infrastructure operators and organizations maintaining widely used technology platforms. More than 650 partners are participating globally, with strict operational requirements including limited access for security teams and multifactor authentication. AI is moving beyond threat detection toward autonomous vulnerability remediation, while controlled access remains central to managing its dual-use risks.
 
ToxicPanda 2.0 expands Android banking malware attacks
Zimperium, a mobile security company, has identified ToxicPanda 2.0, an updated Android banking Trojan with significantly expanded capabilities and global reach. The malware supports 167 remote commands and can target 349 banking, financial, e-wallet and cryptocurrency applications across 16 countries. ToxicPanda 2.0 abuses Android Accessibility Services to remotely control devices, intercept authentication information, steal PINs and credentials, and perform fraudulent transactions. It can also exploit Android Wireless Debugging to obtain shell-level access and extract lock-screen credentials. The malware uses overlay attacks to imitate legitimate applications and deceive users into entering sensitive information. Researchers also found that ToxicPanda 2.0 has adopted AWS-hosted infrastructure and continues to evolve its command-and-control capabilities, indicating an ongoing effort to expand its reach and operational flexibility.