Last year was undeniably the year of Generative AI (hype). According to Muhammad Alam, President and Chief Product Officer of SAP’s Intelligent Spend and Business Network, “generative AI was like a bolt of lightning that compelled every business and technology leader to sit up and take notice.”
Now the first flurry of investment and breathless hype seems to be settling down. This leaves organisations interested in the potential applications of generative AI to try and figure out what that might actually look like.
AI is only as good as its data
Alam notes that those looking to adopt will “see firsthand that AI is only as effective as the quality and availability of data.”
One class of organisation with access to vast amounts of data is government and other public sector entities. This class of organisation also cursed with a host of problems ranging from inefficient organisation to compliance and shrinking budgets.
As researchers at Deloitte noted in a recent report, “Government procurement professionals need help.” According to experts, generative AI could hold the solution to these problems.
“If government is to achieve the ambitious aims that the public expects, procurement professionals need tools to process large volumes of data with precision and with attention to the unique circumstances of every contracting action,” the report adds.
Public procurement under strain
Procurement is currenting caught in a rising tide of complexity. Over the past 10 years, government spending has grown around 4.5% each year in the US. Over the same period, the total number of contracting actions has increased by more than 22% each year.
Public procurement is drowning under layers of complexity which are growing faster than the public sector procurement workforce.
At the same time, workforce headcounts are being stymied by budgetary concerns and an industry wide talent shortage. This is being exacerbated by the fact the private sector has and always will pay better. Therefore, the amount of work being done by individual public sector procurement staff is rising.
In 2022, for every Federal contracting officer, an average of 2,000 contracting actions had to be executed per year. Comparing that number to the 300 actions per year in 2013 reveals the scope of the problem. If government procurement departments are going to avoid buckling under this growing strain, technology in the form of automation, advanced analytics, and other potential generative AI applications could have a role to play.
Generative AI still has optical (and ethical) problems
Generative AI is a somewhat nebulous umbrella term. It is often conflated with its most public-facing examples: Chat-GPT and image generators like Midjourney. This is why there appears to be a disonnect between what generative AI promises and what it delivers.
These models are less effective than humans at doing a lot of things like making art, generating movie scripts, and accurately retrieving or summarising information from the internet, etc. In addition to untrustworthy results and “hallucinations”, large language model AIs and imager generators also have significant ethical issues baked in. These stem from the uncredited work by writers and artists used to train these models.
Applications for a generative Ai layer in public procurement
However, this doesn’t mean that generative AI is a useless or irredeemably immoral technology. Under the right regulatory constraints and in the correct context, Generative AI can create a vital unifying layer between several other pieces of technology. (Obviously more AI regulation is something to which tech industry people seem more reflexively averse than ipecac).
However, generative AI shows promise as an intermediary layers between automation tools, big data analytics, and e-procurement platforms. Deployed correctly, it could alleviate the growing complexity that plagues public sector procurement.
Deloitte’s researchers note, similarly, that “the emergence of gen AI has put a missing puzzle piece on the table that can allow several different types of tools to fall into place. Because gen AI works differently than previous generations of AI, it has different strengths and weaknesses. While gen AI can do things that traditional machine learning (ML) cannot, such as creating new text or images, it can occasionally struggle with accuracy in ways that traditional ML does not. Similarly, all forms of AI can exceed a human’s ability to handle large volumes of data, but humans naturally excel at tasks that strain AI, such as highly variable or social tasks.”
By contrast, tasks like documenting and reporting are hugely time consuming.
“From using gen AI to generate documents and reports to having ML produce demand forecasts, AI can help reduce the time needed to create and process procurement request documents such as market research reports, statements of work, and purchase requests,” notes the report.
