Subhalata Pradhan, a Gram Vikas fieldworker, talks to Raja Pradhan about the chatbot and addresses concerns over sharing his details. Subhalata Pradhan, a Gram Vikas fieldworker, talks to Raja Pradhan about the chatbot and addresses concerns over sharing his details. India's scattered workforce: the chatbot keeping families in touch during emergencies Covid exposed the lack of data on the country's 140 million mobile migrant workers, but a new project in Odisha is helping to fill in the gaps Mon 16 Mar 2026 02.00 EDTLast modified on Mon 16 Mar 2026 02.03 EDT Raja Pradhan is sitting cross-legged, scrolling on his phone in his village in eastern India when a green WhatsApp chat bubble pops up on the screen. Are you going outside for work? He reads the message twice, unsure whether to respond.

Representational bias in large language models (LLMs) has predominantly been measured through single-response interactions and has focused on Global North-centric identities like race and gender. We expand on that research by conducting a systematic audit of GPT -4 Turbo to reveal how deeply encoded representational biases are and how they extend to less-explored dimensions of identity. We prompt GPT - 4 Turbo to generate over 7,200 stories about significant life events (such as weddings) in India, using prompts designed to encourage diversity to varying extents. Comparing the diversity of religious and caste representation in the outputs against the actual population distribution in India as recorded in census data, we quantify the presence and "stickiness" of representational bias in the LLM for religion and caste. We find that GPT -4 responses consistently overrepresent culturally dominant groups far beyond their statistical representation, despite prompts intended to encourage representational diversity. Our findings also suggest that representational bias in LLMs has a winner-take-all quality that is more biased than the likely distribution bias in their training data, and repeated prompt-based nudges have limited and inconsistent efficacy in dislodging these biases. These results suggest that diversifying training data alone may not be sufficient to correct LLM bias, highlighting the need for more fundamental changes in model development.

Analytics leader SAS continues to innovate by making it easier to access its powerful, cloud-native SAS Viya platform. Recent product developments help customers more efficiently democratize analytics throughout their organizations while seamlessly managing analytic workloads and building SAS into a variety of applications. "Among analytics providers, SAS is a recognized leader for reimagining the way analytics and machine learning are consumed so users realize optimal results that solve business challenges faster with more accuracy," said Ritu Jyoti, Program Vice President of Artificial Intelligence Research at IDC. "This iteration of the SAS Viya platform gives users at all levels of an organization an agile, powerful advanced machine learning engine to help master the complex data that's in cloud environments." Recognizing that so many organizations have struggled with the global pandemic and its effects, SAS CEO and founder Jim Goodnight says SAS Viya can help in the toughest times. "SAS Viya helps organizations see their data, customers and operations in new ways that encourage confident decision making," said Goodnight.

A global virtual conference aimed at promoting the use of artificial intelligence (AI) and machine learning (ML) for the development of Odisha and advancement of Odia language in the digital era is being organized on 4th October Sunday, on the virtual meeting platform Zoom. The conference, called Odias in ML Conference, is being organized by a group of Odias, also called Odias in ML, with a shared interest in AI and ML. The conference also aims to showcase career and entrepreneurship opportunities in AI and ML for Odias across the world. This first-of-its-kind conference will see participation of a multitude of stakeholders including technologists, researchers, academicians, business executives, entrepreneurs, policymakers, linguists, language activists, media persons and community leaders, all with a commitment to AI and machine learning. The speakers, all Odias based across three continents, will come together to brainstorm how the opportunities created by these emerging technologies can be leveraged effectively to propel the next phase of growth for Odisha and Odias.

BHUBANESWAR: In its bid to ensure greater fiscal accountability, effective resources management and increased productivity, the state government is set to go for integration of artificial intelligence (AI) in the financial auditing mechanism for its organisations in a big way. The Directorate of Local Fund Audit (DLFA) has started work on an AI project that is aimed at reshaping organisational accountability and bringing about greater trust on Government institutions. It is already seeking advisory services from experts for the successful implementation of the technology infusion in local fund audit automation. The Government has appointed Professor (Information Systems) of XIMB Sanjay Mohapatra as Advisor to the AI implementation committee of DLFA. He will assist LFA to develop a comprehensive strategy to analyze processes and develop implementation plans for AI-enabled automation and advise on available options and capabilities along with skilling of officials As per the project, the audit of organisations will be conducted using various components of AI like machine learning, deep learning, natural language processing and computer vision.

The year 2019 has been one of the worst years in the history for India when it comes to natural disasters. More than 13 Indian states (Kerala, Gujarat, Karnataka, Maharashtra, Madhya Pradesh, Andhra Pradesh, Odisha) have been affected due to floods. Government reports state that this year has been witness to the heaviest monsoon rains to wreck havoc in India in the last 25 years. More than 1,600 people have been killed with millions of people losing their homes and their livelihood. While this year has been devastating, every year, the same pattern repeats itself.

Robot restaurants have been popping up all over the globe. While most are located in China and Japan, India has also been getting in on the action. We recently reported on a restaurant in Kerala turning to bots to bring in customers. Now the eastern part of the country is getting its own robot restaurant. It features two humanoid servers named Champa and Chameli.

Every year during the monsoon period, India faces a deluge of floods. Almost every year, it is common to see one or two Indian states being badly affected by floods. Most recently, in Bihar, more than 1,400 villages in 15 districts were marooned due to flooding. In 2019 alone, more than 13 Indian states (Kerala, Gujarat, Karnataka, Maharashtra, Madhya Pradesh, Andhra Pradesh, Odisha) have been affected due to floods. Government reports state that this year has been witness to the heaviest monsoon rains to wreck havoc in India in the last 25 years.

Every year during the monsoon period, India faces a deluge of floods. Almost every year, it is common to see one or two Indian states being badly affected by floods. Most recently, in Bihar, more than 1,400 villages in 15 districts were marooned due to flooding. In 2019 alone, more than 13 Indian states (Kerala, Gujarat, Karnataka, Maharashtra, Madhya Pradesh, Andhra Pradesh, Odisha) have been affected due to floods. Government reports state that this year has been witness to the heaviest monsoon rains to wreck havoc in India in the last 25 years.

Now robot will take orders and serve food replacing the waiters at a'smart' restaurant in Odisha's Bhubaneswar. It is said to be the first restaurant in eastern India where two indigenously developed robots namely'Champa' and'Chameli' are engaged in serving food to the customers. They have voice operated system to greet the customers and wish them. In fact they can speak'Odia' language too. This restaurant has been started by one Odiya Engineer.