GSA Issues Interim AI Acquisition Policy for Federal Contracts

What happened

The U.S. General Services Administration has issued an artificial intelligence acquisition policy in the form of a regulation deviation, allowing federal agencies to apply it to new contracts immediately rather than waiting for a formal rulemaking process to conclude. According to FedScoop, the deviation reflects some of the public comments GSA received as it works toward a permanent AI acquisition rule.

By using a deviation rather than waiting for the standard regulatory process, GSA is able to give contracting officers and agencies interim guidance on how to acquire AI systems and services now, while the broader rulemaking continues in parallel.

Why it matters

Government procurement rules move slowly, and AI capabilities are moving fast — a mismatch that has left many federal contracting officers without clear guidance on how to buy AI responsibly. By issuing a usable deviation now, GSA is signalling that it wants agencies acquiring AI tools to have a documented policy framework to work from today, rather than defaulting to ad hoc interpretations of older procurement rules that were never written with AI in mind.

For digital transformation leaders across government, this is a practical unlock: it reduces uncertainty for agencies that have been hesitant to procure AI tools absent explicit rules, and it gives vendors a clearer target to design their contracting approach around. It also sets a template other public-sector bodies — in the U.S. and internationally — may watch closely as they face the same tension between regulatory caution and the pace of AI adoption.

The Renascence take

What's notable here isn't the content of the policy itself but the mechanism GSA chose to deliver it.

Most public-sector AI debates fixate on what the rules should eventually say. GSA's move is really a lesson in sequencing: an interim, usable policy beats a perfect one that arrives too late to matter. Organisations — public or private — that are stalling AI adoption while waiting for "final" governance frameworks should take note. The behavioral reality is that uncertainty, not risk itself, is what freezes decision-makers; issuing workable interim guidance, even if it's revised later, unlocks action faster than waiting for consensus. The real test for GSA now is whether this deviation actually gets used, or simply sits alongside the rulemaking as a parallel track agencies ignore.