Eigen RadarAI
Analysis

India Mobile Congress closes with calls for practical AI evaluation

India Mobile Congress ended its four-day program with discussions on agricultural AI, evaluation frameworks and autonomous agents for 5G networks. Participants called for fairness, transparency and robustness in testing, alongside data quality and farmers’ control over information. The closing sessions focused on how standards and practical development could support applications beyond isolated pilot projects.

Artificial Intelligence··Night
Telecom test equipment and an antenna connected by cables in a bright conference area.

The final day linked agricultural AI and network development

India Mobile Congress, a telecommunications and technology gathering in India, concluded its four-day program on 10 October. Its final day included a meeting on AI for agri-food systems held with the International Telecommunication Union, or ITU. Engineers at the ITU Build-A-Thon worked on autonomous AI agents for managing 5G networks.[1], [2]

AI evaluation covers bias and difficult operating conditions

Devendra Singh of India’s Telecommunication Engineering Centre, a government body working on telecommunications technology, called for structured evaluation frameworks covering fairness, transparency and robustness. He identified possible sources of bias in uncurated training data, outliers, model selection and deployment environments. He described testing across risk tiers with domain-specific metrics and empirical checks under adverse conditions. Reliable operation and measurable outputs were part of his account of robustness.[1]

Farmers’ needs and control over data shape the standards discussion

ITU official Bilel Jamoussi emphasized solving real problems for smallholder farmers and using international standards to avoid isolated pilots. Syed Tausif Abbas called for tangible improvements in livelihoods, agricultural productivity, supply chains and food security, with attention to data quality and privacy.[1], [2]

Jan Redlich, technology chief at Bonn Systems, described a reference architecture using shared semantics and open interfaces while preserving control over data. Debasish Mahalik of the Bureau of Indian Standards discussed data quality, security and interoperability across the system. The closing-day statements concerned evaluation priorities and standards; they did not announce a binding standard or a newly released AI product.[1]

References

  1. News sourceTelecomTalkIndia Mobile Congress closes with discussions on AI evaluation and network agents↩1↩2↩3↩4
  2. News sourceWebIndia123 / ANIIndia Mobile Congress closes with discussions on AI evaluation and network agents↩1↩2