Trust is key, tech fund needs more guardrails
The government's fund supporting deep-tech sectors like AI, space and quantum technology is promising, yet private sector research and development spending remains minimal, risking the nation falling behind in the tech race without long-term, patient funding like venture capital. However, the collaboration between public funds and private networks is a complex issue, as taxpayers, not shareholders, bear the risk.
To ensure fairness, the paper investigates the Research, Development and Innovation Fund's first round, which approved Rs 2,192 crore in soft loans to 22 private companies chosen by a 12-member panel, 11 from private equity and technology, and one non-voting government representative. While drawing experts from the private sector is a wise decision, 15 of the 22 selected firms have investment ties to seven of those panel members, and nine trace back to the committee's chairman.
This raises concerns about the process's fairness, with officials claiming decisions were made on merit by unconflicted members. Although disclosure and recusal are standard conflict management tools, they only address voting conflicts, not the informational and reputational advantages of panel membership or the pull of investors who also sit on committees controlling public capital access.
In an economy with concentrated ownership and capital, a public fund for deep-tech cannot rely solely on the good faith of its selectors. The solution lies in balancing innovation instincts with compliance discipline, expanding and rotating the selectors' pool to prevent disproportionate influence, incorporating outside academic and scientific evaluators for initial screening, separating technical assessment from financial interest, and publishing conflict disclosures and recusal records as mandated practice. A fund built on public money must earn trust through its architecture, not just its assertions.
Written by urgent.news from The Indian Express's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.