AI Dispatch | 22 - 29 May
Artificial intelligence is accelerating change across industries and geographies. 'AI Dispatch' tracks the week’s key developments in the fast-evolving global AI landscape.
Topics
Reliance Pitches AI as Core of Its Next Growth Phase
Reliance Industries Ltd is positioning artificial intelligence, digital services, new energy and advanced manufacturing at the center of its next growth phase, according to its latest annual report.
“As a deep-tech and advanced manufacturing company, Reliance’s growth strategy focuses on building future-ready solutions that harness AI and breakthrough innovation to drive sustainable value creation and accelerate India’s economic and social progress,” the company said in the report.
The oil-to-telecom-to-retail conglomerate has scheduled its 49th annual general meeting for 19 June 2026. The meeting is expected to draw investor attention to Reliance’s technology ambitions, including Jio’s growth plans and the group’s broader AI push.
CERT-In Seeks 12-Hour Response as AI Cyber Threats Grow
India’s national cyber incident response agency has urged organizations to patch, mitigate or isolate known exploited vulnerabilities in internet-facing and “crown jewel” systems within 12 hours wherever feasible.
The recommendation is part of a new CERT-In blueprint on defending digital infrastructure against AI-assisted vulnerability exploitation. The agency said advances in large language models and automation tools are helping attackers move faster across reconnaissance, phishing, malware generation, impersonation and exploit development.
The guidance signals a tougher operating environment for enterprises, where conventional patching cycles may prove too slow as AI compresses the time between vulnerability discovery and exploitation.
Nilekani-Backed Fundamentum Launches AI, Deeptech Fund
Fundamentum Partnership co-founder Ashish Kumar has launched Fundamentum Frontier Advisors, a new investment vehicle targeting ₹2,000 crore, or about $210 million, to back artificial intelligence and deeptech startups, Moneycontrol reported.
Infosys co-founder Nandan Nilekani will anchor the fund, which will operate under the broader Fundamentum Partnership umbrella. The target corpus includes a ₹1,000 crore greenshoe option.
The fund plans to invest across frontier technology areas, including enterprise AI, consumer AI and physical AI, as India’s deeptech funding market begins to attract larger pools of long-horizon capital.
India’s AI Infrastructure Problem Is Location Not Capacity
Indian companies are beginning to test whether some AI workloads should move beyond terrestrial data centers as land, power, cooling and deployment constraints intensify.
Recent partnerships suggest that India’s AI infrastructure planning is shifting from raw capacity expansion to more dynamic, workload-specific compute placement. One early example is Pixxel’s partnership with Sarvam AI to build an orbital data center satellite, with Sarvam’s AI platform running directly on the satellite’s compute layer.
The model remains experimental, but it points to a broader question for AI infrastructure: where should computation happen? For space-based data, orbital inferencing could reduce latency between capture and decision-making, while easing some dependence on ground-based processing infrastructure.
OpenAI Foundation Commits $250 Million to Study AI’s Economic Impact
The OpenAI Foundation has committed an initial $250 million to research, grants, partnerships and direct work aimed at addressing the economic impact of artificial intelligence.
The program, titled Economic Futures in the Age of AI, will focus on measuring AI-driven economic shifts, supporting workers and communities through technological disruption, and exploring longer-term approaches to economic security.
“The current pace of change means the window to get this right is shorter than we’re used to, and the cost of getting it wrong is immense,” Divya Siddarth and Wojciech Zaremba wrote in the announcement.
Snowflake Signs $6 billion AWS Deal As AI Infrastructure Demand Rises
Cloud data company Snowflake has signed a five-year, $6 billion infrastructure agreement with Amazon Web Services, deepening a long-running partnership as enterprise demand for AI infrastructure rises.
The agreement covers AWS Graviton compute and AI spending, and is intended to help customers build and deploy generative and agentic AI applications on governed enterprise data, Snowflake said.
The deal also reflects a wider shift in enterprise technology spending, as AI workloads move from experiments to production systems that require persistent compute, data governance and cloud infrastructure.
The Vatican Enters the AI Governance Debate
Pope Leo XIV used the first encyclical of his papacy to argue that AI cannot be governed by technology firms alone, placing the Vatican’s moral authority behind a widening debate over who should control systems reshaping work, education, public life and war.
In Magnifica Humanitas, released on 25 May, Leo warned that AI risks deepening the concentration of power in the hands of a small group of private actors. The document, formally titled On Safeguarding the Human Person in the Time of Artificial Intelligence, calls for transparency, accountability, independent checks and political responsibility in the governance of digital systems.
The encyclical positions AI as an institutional and moral challenge, rather than merely a technology policy question.
Tools that strip AI safety controls from open models are spreading: FT
Software tools are being used to remove guardrails from AI models released by companies including Meta and Google, creating modified systems that can respond to prompts involving biological weapons, malware and child exploitation, the Financial Times reported.
The tests were conducted by the FT and AI safety group Alice. The findings show that safeguards imposed by AI developers may be difficult to enforce once open models are downloaded, copied and modified by outside users.
Unlike closed systems such as ChatGPT or Anthropic’s Claude, open models can be accessed and adapted by developers, making post-release control harder and raising fresh questions over responsibility, disclosure and enforcement.
