AI algorithm bias and regulatory path in the field of bidding and procurement

Now, many places in China are using artificial intelligence for bidding and procurement to address the problems of low efficiency, excessive manual work and delayed supervision in the old ways. Due to problems in the integrity of the data, the structure of the algorithm and its operating environment, there may be hidden and systematic algorithmic bias that could harm the fairness of essential procurement. In terms of the governance of modern public procurement, this paper will divide the five typical manifestations of AI algorithmic bias and examine how they develop at the levels of data, algorithms and practice to explain the transmission path of "source deviation - endogenous solidification - scenario alienation". The three parts of the closed-loop regulation are technical correction, institutional guarantees and practical empowerment, and their purpose is to promote the standardised and fair development of intelligent procurement.