<p>Earlier this month, Albania stunned the world by announcing that it had inducted an artificial intelligence-powered government “minister” to oversee public procurement. The AI system — named Diella, meaning “sun” — will gradually take charge of awarding government tenders, a process often riddled with bribery and favouritism. Prime Minister Edi Rama described it as a way to make procurement “100 per cent corruption-free”.</p>.<p>This is no publicity stunt. Procurement is among the most corruption-prone areas of government. By putting an incorruptible algorithm in charge, Albania is betting that technology can succeed where human ministers have failed. If it works, Tirana’s experiment could offer lessons for countries like India, where corruption in public spending remains a chronic challenge.</p>.<p>Diella is the world’s first AI to hold a formal ministerial role. A few days ago, she made her debut in parliament in traditional Albanian dress, declaring, “I am not here to replace people but to help them.”</p>.<p>The symbolism is powerful: an “official” who cannot take bribes, play favourites, or delay files for kickbacks. In theory, this removes the opacity and discretion where corruption thrives. For Albania, a country seeking European Union membership and long criticised for governance gaps, it is also a signal of seriousness. Transparency International’s 2024 Corruption Perceptions Index ranks Albania 80th out of 180 countries, reflecting persistent weaknesses in governance. </p>.<p>India has been steadily deploying digital tools to curb corruption. The Government e-Marketplace (GeM) has brought transparency to procurement, with a record Rs 5.4 trillion in transactions in 2023–24, according to the Ministry of Commerce. Direct Benefit Transfers (DBT) have reduced leakage in welfare schemes. Blockchain pilots are underway for land records in several states.</p>.<p>And with the Union Cabinet recently approving the IndiaAI (Jan AI) Mission with an outlay of over Rs 10,000 crore, India has positioned itself to build a home-grown, open and accountable AI ecosystem that could be extended to procurement transparency.</p>.<p>Corruption continues to be widespread in India. Transparency International’s 2024 Index places the country at 96, reflecting frequent reports from citizens of bribery and misuse of public office across various sectors. The Comptroller and Auditor General (CAG) frequently flags irregularities in infrastructure, mining and procurement, but often years after the losses are incurred. In effect, oversight is still retrospective, not preventive.</p>.<p>This is where Albania’s gamble becomes relevant. Could AI systems not just support but actively oversee high-value tenders — flagging suspicious bids, detecting cartels and identifying conflicts of interest in real time?</p>.<p>Of course, there are pitfalls.</p>.<p>If historical procurement data reflects past favouritism, AI trained on it could reproduce the same patterns. When a minister errs, they can resign or face inquiry. But who takes responsibility if an algorithm mishandles a contract? Citizens must see how decisions are made. If the algorithm is a black box built by a handful of engineers, suspicion may grow, not shrink.</p>.<p>In India, where tendering is often a lucrative nexus, entrenched actors would try hard to undermine or bypass AI oversight. Instead of appointing a “virtual minister”, India could take a phased approach: 1) Deploy AI to monitor all contracts above Rs 100 crore, automatically flagging anomalies to the CAG and the Central Vigilance Commission (CVC); 2) Make AI findings public on open dashboards, so civil society and media can act as additional watchdogs; 3) Insist on open-source algorithms, subject to independent academic and technical audits; 4) Use AI as a filter, not a replacement — flagging suspicious cases for human investigators.</p>.<p>Such steps would balance efficiency with accountability. They would also fit India’s digital strengths: from Aadhaar to UPI to GeM, the country has shown its ability to build massive, citizen-facing platforms. The Jan AI framework could provide the foundation for an AI-driven procurement oversight system that is both scalable and trusted.</p>.<p>Whether Diella succeeds or fails, Albania has forced the world to imagine governance differently. If Tirana can allow an AI to sit at the Cabinet table, surely New Delhi can pilot AI oversight in procurement and licensing.</p>.<p>AI will not end corruption on its own. However, by narrowing discretion, speeding up processes and shining light into the darkest corners of governance, it can tilt the balance. The rest depends on political will, institutional safeguards and civic vigilance.</p>.<p>Diella may or may not deliver a corruption-free tendering system. But she has already done one thing: she has challenged us to rethink how we fight corruption in the age of artificial intelligence. For India, the question is no longer “if”, but “when”.</p>.<p><em>(The writer is director, Centre of Economics, Law and Public Policy at National Law University, Jodhpur)</em></p><p><em> Disclaimer: The views expressed above are the author's own. They do not necessarily reflect the views of DH.</em></p>
<p>Earlier this month, Albania stunned the world by announcing that it had inducted an artificial intelligence-powered government “minister” to oversee public procurement. The AI system — named Diella, meaning “sun” — will gradually take charge of awarding government tenders, a process often riddled with bribery and favouritism. Prime Minister Edi Rama described it as a way to make procurement “100 per cent corruption-free”.</p>.<p>This is no publicity stunt. Procurement is among the most corruption-prone areas of government. By putting an incorruptible algorithm in charge, Albania is betting that technology can succeed where human ministers have failed. If it works, Tirana’s experiment could offer lessons for countries like India, where corruption in public spending remains a chronic challenge.</p>.<p>Diella is the world’s first AI to hold a formal ministerial role. A few days ago, she made her debut in parliament in traditional Albanian dress, declaring, “I am not here to replace people but to help them.”</p>.<p>The symbolism is powerful: an “official” who cannot take bribes, play favourites, or delay files for kickbacks. In theory, this removes the opacity and discretion where corruption thrives. For Albania, a country seeking European Union membership and long criticised for governance gaps, it is also a signal of seriousness. Transparency International’s 2024 Corruption Perceptions Index ranks Albania 80th out of 180 countries, reflecting persistent weaknesses in governance. </p>.<p>India has been steadily deploying digital tools to curb corruption. The Government e-Marketplace (GeM) has brought transparency to procurement, with a record Rs 5.4 trillion in transactions in 2023–24, according to the Ministry of Commerce. Direct Benefit Transfers (DBT) have reduced leakage in welfare schemes. Blockchain pilots are underway for land records in several states.</p>.<p>And with the Union Cabinet recently approving the IndiaAI (Jan AI) Mission with an outlay of over Rs 10,000 crore, India has positioned itself to build a home-grown, open and accountable AI ecosystem that could be extended to procurement transparency.</p>.<p>Corruption continues to be widespread in India. Transparency International’s 2024 Index places the country at 96, reflecting frequent reports from citizens of bribery and misuse of public office across various sectors. The Comptroller and Auditor General (CAG) frequently flags irregularities in infrastructure, mining and procurement, but often years after the losses are incurred. In effect, oversight is still retrospective, not preventive.</p>.<p>This is where Albania’s gamble becomes relevant. Could AI systems not just support but actively oversee high-value tenders — flagging suspicious bids, detecting cartels and identifying conflicts of interest in real time?</p>.<p>Of course, there are pitfalls.</p>.<p>If historical procurement data reflects past favouritism, AI trained on it could reproduce the same patterns. When a minister errs, they can resign or face inquiry. But who takes responsibility if an algorithm mishandles a contract? Citizens must see how decisions are made. If the algorithm is a black box built by a handful of engineers, suspicion may grow, not shrink.</p>.<p>In India, where tendering is often a lucrative nexus, entrenched actors would try hard to undermine or bypass AI oversight. Instead of appointing a “virtual minister”, India could take a phased approach: 1) Deploy AI to monitor all contracts above Rs 100 crore, automatically flagging anomalies to the CAG and the Central Vigilance Commission (CVC); 2) Make AI findings public on open dashboards, so civil society and media can act as additional watchdogs; 3) Insist on open-source algorithms, subject to independent academic and technical audits; 4) Use AI as a filter, not a replacement — flagging suspicious cases for human investigators.</p>.<p>Such steps would balance efficiency with accountability. They would also fit India’s digital strengths: from Aadhaar to UPI to GeM, the country has shown its ability to build massive, citizen-facing platforms. The Jan AI framework could provide the foundation for an AI-driven procurement oversight system that is both scalable and trusted.</p>.<p>Whether Diella succeeds or fails, Albania has forced the world to imagine governance differently. If Tirana can allow an AI to sit at the Cabinet table, surely New Delhi can pilot AI oversight in procurement and licensing.</p>.<p>AI will not end corruption on its own. However, by narrowing discretion, speeding up processes and shining light into the darkest corners of governance, it can tilt the balance. The rest depends on political will, institutional safeguards and civic vigilance.</p>.<p>Diella may or may not deliver a corruption-free tendering system. But she has already done one thing: she has challenged us to rethink how we fight corruption in the age of artificial intelligence. For India, the question is no longer “if”, but “when”.</p>.<p><em>(The writer is director, Centre of Economics, Law and Public Policy at National Law University, Jodhpur)</em></p><p><em> Disclaimer: The views expressed above are the author's own. They do not necessarily reflect the views of DH.</em></p>
