Using AI to assist in government procurement

By Grace Edinger, Senior Manager of the Restoration Economy Center, and Reed Van Beveren, Senior Manager of Technology Policy

Government procurement is a complex process that involves a significant amount of paperwork and communication among various stakeholders. The adoption of artificial intelligence (AI) can help streamline this process and improve the efficiency of government procurement.

One application of AI in government procurement is the use of predictive analytics to identify potential risks and opportunities in the procurement process. This can help government staff to make informed decisions about procurement contracts and reduce the likelihood of fraud or errors. Predictive analytics can also be used to identify areas where cost savings can be achieved, such as by identifying suppliers who offer the best value for money.

Another application of AI in government procurement is the use of chatbots to handle routine inquiries from suppliers and contractors. This can help to free up government staff to focus on more complex tasks, such as evaluating bids and negotiating contracts. Chatbots can also provide real-time updates to suppliers about the status of their bids or contracts, which can help to improve communication and reduce the likelihood of disputes.

AI can also be used to automate the procurement process, from the initial request for proposal (RFP) to the awarding of contracts. This can help to reduce the time and cost associated with manual procurement processes, and can help to ensure that the procurement process is fair and transparent. Automated procurement can also help to improve the accuracy of procurement data, which can help to inform future procurement decisions.

Overall, the adoption of AI in government procurement can help to improve the efficiency, transparency, and accuracy of the procurement process. By reducing the administrative burden on government staff and providing real-time updates to suppliers, AI can help to improve communication and reduce the likelihood of errors and disputes. Furthermore, the use of predictive analytics and automated procurement can help to identify cost savings and ensure that the procurement process is fair and transparent.

***

Everything written up until this point was done by ChatGPT. The prompt we used was “write a 4 paragraph blog that explains the applications of AI to government procurement, including benefits to government staff.” The response is a good summary of the general promise of AI for government procurement, but how could AI applied to procurement help speed up environmental progress specifically? Two different EPIC teams, the Restoration Economy Center and the Technology Team, share their thoughts below: 

Grace Edinger, Senior Manager of the Restoration Economy Center:
As someone who works on innovative contract structures for ecological outcomes, I hear time and again that state government staff aren’t comfortable with stepping outside of the status quo, largely because they don’t have the internal capacity or knowledge to take it on. 

Even with clear legal authority, agency staff across the country have told us that contract templates and examples would be immensely helpful. With the direction that AI technology is headed, I see a future where agency staff could essentially ask for AI to draft a specific contract, based on their own needs rather than needing to adapt a static template. Not to say that human eyes shouldn’t revise and edit as needed, but the issues of capacity and knowledge could easily be mitigated. 

In order for AI to produce high-quality contracts, it needs high-quality training material. As contract structures evolve and become more prevalent, the body of examples that can be used to train the tech becomes more robust. Rather than individually sharing these examples, we can capitalize on the entrepreneurial spirit of these agencies and states and feed their success into tech that can assist others in following suit. 

Reed Van Beveren, Senior Manager of Technology Policy:
To get the best information for decision-making, environmental agencies need to be able to partner with external technology providers quickly. In practice, there are a number of challenges preventing this that the use of AI could help address. 

For one, information that technology providers need to work with the government, like how to submit proposals and who to contact, is spread out across dozens of websites. AI search and synthesis tools could help tech providers navigate this information, find programs, and contacts to investigate potential partnerships faster. Likewise, it can be challenging for government agencies to conduct the kind of continuous and proactive market research required to adopt new technology effectively. AI can help sort through vast sources of information and look for patterns depending on the specific goals of the government. The Department of Homeland Security is starting to evaluate AI’s potential in this area.

Delays in the procurement process is one major reason why some tech providers avoid working with federal agencies. There is often a mismatch between the federal procurement timelines and the short development and funding cycles of innovative technology providers. Status updates on proposals and how close government agencies are to making a decision are few and far between, especially when agencies are understaffed. AI has the potential to automate or assist contracting officials with first drafts of contracts to help shorten that time and allow for more automated and frequent communication to keep everyone engaged in the process. The Department of Defense is leading the way with Acqbot, a tool that can help draft a full contract quickly. 

Conclusion
The use of AI assisted procurement processes is an emerging opportunity to push further and faster, and help government staff get the most out of overwhelming investments in nature through the IIJA and IRA. While it may not be ready for use in all situations, now is the time for environmental agencies at all levels to start thinking about how advances in AI can enable better investment in environmental tech and outcomes in the months and years ahead.

Previous
Previous

EPIC Featured: Ecobite podcast about innovation and wetlands.

Next
Next

A New Job, A Transition, and A Commitment to Planet Earth