New Delhi, Nov. 6 -- This week's enterprise technology developments reflect a sharper focus on putting AI to work while addressing the risks and infrastructure demands that come with wider adoption. From tighter regulatory scrutiny of AI systems and investments in domestic computing capacity to enterprise-scale agentic AI deployments and 5G-enabled railway safety, the developments signal how technology is reshaping business operations and critical infrastructure. Here's a roundup of the key developments.

1. CERT-In issues AI security audit guidelines for enterprises India's cybersecurity agency CERT-In has issued AI security audit guidelines, requiring reportable AI-related security incidents to be disclosed within six hours, in line with its existing cybersecurity directions. The framework classifies AI systems into four risk categories: low, moderate, high and critical, based on factors including autonomy, data sensitivity and potential impact. It outlines nine audit domains covering AI governance, data privacy, model risk management, bias, explainability and security. Organisations are also expected to maintain traceability of training data, model configurations, prompts and system changes, alongside retaining security logs for at least 180 days.

Earlier this week, the agency flagged that attackers are using AI tools to automate intrusions into non-banking financial companies, fintech firms, payment aggregators and cryptocurrency exchanges. Separately, CERT-In has secured controlled access to Anthropic's Claude Mythos model to test its ability to identify software vulnerabilities.

2. Ericsson, HBL integrate Kavach 4.0 with 5G network for railway safety Ericsson and HBL have integrated India's indigenous Kavach 4.0 automatic train protection system with a 5G Standalone (SA) network environment at HBL's research and development laboratory in Hyderabad. The integration validates interoperability between the railway safety system and 5G technology, opening up possibilities for next-generation railway communications. Kavach 4.0, which is SIL-4 certified, is designed to prevent collisions and Signal Passing at Danger (SPAD), while supporting safer train operations through continuous supervision. The companies are evaluating how 5G SA could support signalling, mission-critical communication, onboard camera surveillance and predictive maintenance. The development brings telecom infrastructure and industrial safety technology together, with potential applications in India's railway modernisation programme.

3. IndiaAI Mission unveils indigenous models for enterprise use The IndiaAI Mission has released three indigenous foundation models: EKA-1, developed by Soket Labs, and Dhir 2B and Dhir 5B, developed by ZenteiQ. EKA-1 is a 26.3-billion-parameter model trained on approximately 7.5 trillion tokens, supporting all 22 scheduled Indian languages and English. It is designed for mathematical reasoning, coding and agentic applications. The Dhir models are open-source multilingual assistants supporting English and 10 Indian languages, with options for secure, on-premise and offline deployment. Their intended applications include document-based automation, knowledge retrieval and enterprise services across sensitive sectors. The launch adds to India's efforts to develop domestic AI capabilities and expand access to models suited to local language and data requirements.

4. LTM deploys 1,500 AI agents with ServiceNow LTM and ServiceNow have expanded their strategic partnership to help enterprises scale agentic artificial intelligence (AI) across business functions, moving beyond isolated deployments towards organisation-wide adoption. The collaboration will combine ServiceNow's AI Platform with LTM's engineering, domain and transformation expertise to develop joint solutions and go-to-market programmes for more than 700 shared customers. As part of its internal transformation, LTM has deployed 1,500 ServiceNow AI agents across its operations, where they work alongside employees across cybersecurity, human resources, IT and project delivery. The company is also using ServiceNow's Super Intelligence Control Tower to monitor its AI assets and associated risks, expecting up to an 80% improvement in AI asset management efficiency. The deployment will serve as a reference for helping other enterprises scale agentic AI with governance and accountability.

5. India's data centre capacity to grow 5x by 2030 India's data centre capacity is expected to increase five-fold to approximately 7.5 GW by March 2030, attracting investments of more than Rs 3 trillion over the next four years, according to Crisil Ratings. The expansion is being driven by generative AI computing demand, enterprise digitalisation and rising data consumption. Nearly half of the upcoming capacity is already committed under long-term contracts, providing revenue visibility for developers. However, power availability, grid connectivity and project execution could constrain the pace of expansion. Visakhapatnam, Mumbai and Hyderabad are expected to account for a significant share of upcoming capacity. The findings underline the infrastructure requirements of India's AI and cloud ambitions, particularly the need for reliable power and facilities capable of supporting higher-density computing workloads.

6. NPCI sets up tech subsidiary for AI and blockchain The National Payments Corporation of India (NPCI) has established NPCI Tech Solutions Ltd, a wholly owned subsidiary that will serve as its technology and innovation arm. The new entity will focus on artificial intelligence, blockchain, digital financial infrastructure and global capability centre services. It will also provide technology services for initiatives such as eKYC Setu and the Unified Ticketing Platform. NPCI's chief technology officer Vishal Kanvaty has been appointed managing director and chief executive of the subsidiary. The move is aimed at building technology capabilities beyond NPCI's core payments mandate and exploring emerging technologies that can support India's digital financial ecosystem. It also signals a broader push by the organisation to develop new technology platforms around digital public infrastructure.

7. Vi partners PwC India for enterprise agentic AI Vodafone Idea (Vi) has partnered with PwC India to establish the Vi Intelligence Lab, an initiative aimed at developing and scaling artificial intelligence solutions for enterprises. The lab will combine Vi's telecom and digital infrastructure with PwC's expertise in business transformation, bringing together AI companies, startups, research institutions and enterprise customers. PwC will set up a dedicated Centre of Excellence team alongside Vi in Mumbai. The initiative will focus on agentic AI, assistive AI, and sovereign and responsible AI, with potential applications in customer engagement, service delivery and cross-functional business processes. The partnership reflects how telecom operators are looking beyond connectivity towards enterprise AI services, as businesses explore automation and increasingly autonomous workflows.

8. AMD's AI-led data centre revenue doubles to $8.5B Advanced Micro Devices (AMD) reported a 97% year-on-year increase in data centre revenue to $8.5 billion in the third quarter of 2026, driven by demand for its EPYC server processors and Instinct graphics processing units. Overall quarterly revenue rose 45% to $13.38 billion, with the data centre business accounting for the largest share. The company forecast fourth-quarter revenue of approximately $17 billion, plus or minus 3%, reflecting continued demand for its computing products. The results underline the growing contribution of AI infrastructure to semiconductor industry revenues as cloud providers and technology companies expand their computing capacity. For Indian enterprises and data centre operators, the performance offers an indication of the global investment cycle shaping AI infrastructure procurement.

9. TRAI proposes AI governance rules for telcos On 2 November 2026, TRAI proposed a framework to make telecom operators accountable for artificial intelligence systems used in their networks, including those developed by third-party vendors. The draft guidelines call for board-approved AI policies, risk-based classification of AI systems, pre-deployment safety assessments and dedicated AI governance committees. Operators would also need to maintain inventories of AI systems, provide human oversight and report serious AI-related incidents to the regulator. The proposals are particularly relevant as telecom companies increasingly use AI for network optimisation, customer services, fraud detection and predictive maintenance. The consultation is open until 27 November 2026.

Published by HT Digital Content Services with permission from MINT CIO CIRCLE.