Georgia's version of the 2024 International Building Code took effect January 1, 2026, adopted with state amendments by the Board of Community Affairs under O.C.G.A. Section 8-2-23. Most of the coverage on it led with the same hook: for the first time in the code's history, Risk Category III and IV buildings in the tornado-prone region now have to be designed for tornado loads. IBC Section 1609.5 and Figure 1609.5 set the trigger; the actual load provisions live in Chapter 32 of ASCE 7-22, a separate standard the IBC points to, not a chapter of the IBC itself. That is a real story. It is also not the change most likely to bite an architecture firm's QC review this year.
The change more likely to bite sits a few sections earlier, in the same chapter of the IBC, in Table 1604.5. It does not come with a press release.
What actually moved
Under the 2024 IBC, every Group I-2 occupancy is now assigned Risk Category IV. Group I-2 covers buildings used for medical care, on a 24-hour basis, of more than five people incapable of self-preservation: hospitals, nursing homes, psychiatric hospitals, detoxification facilities, and foster care facilities, per IBC Section 308.3. Facilities of five or fewer residents fall under a different classification and are not swept in by this change. Under the prior edition, a qualifying Group I-2 facility only landed in Risk Category IV if it provided emergency surgery or emergency treatment. A nursing home or a standalone detox facility without an emergency component could sit at Risk Category II or III. That distinction is gone. If the occupancy is I-2, the risk category is now IV, full stop, according to the 2024 IBC structural changes summary published by Structure Magazine, written by John "Buddy" Showalter, P.E., Senior Staff Engineer of ICC's Consulting Group, and Sandra Hyde, P.E., Managing Director of that same group.
Risk category is not a filing detail. Under ASCE 7-22, it selects which wind speed map a project reads its design wind speed from, and it sets the seismic importance factor used in structural design. Under the 2024 IBC it is also the gate for whether tornado loads apply at all: per IBC Section 1609.5, only Risk Category III and IV buildings in the tornado-prone region shown in Figure 1609.5, roughly the area east of the Continental Divide, get evaluated under ASCE 7-22 Chapter 32, according to both the code text itself and the ICC's own technical writeup by Frank Woeste and Scott Coffman. So a standalone nursing home that used to sit at Risk Category II now sits at Risk Category IV for the sole reason that it is I-2 occupancy, and in an adopting jurisdiction within the tornado-prone region, that reclassification is what pulls it into a tornado load case it would not otherwise reach. A nursing home that already sat at Risk Category III reaches that same load case once the jurisdiction adopts the 2024 code, independent of the I-2 change, because Risk Category III was already in scope for the tornado provision.
Where that number actually lives in a set
Risk category shows up in exactly one or two places in most architectural CD sets: the code data block on the cover or general notes sheet, and a cross reference on the structural general notes. It is rarely something the architectural QC pass re-derives. On a repeat client, it is often carried forward from the last project of the same type, because the team already "knows" what category a nursing home sits at. That habit was harmless for a long time, because the classification rules did not move much between editions. They just moved, for one whole occupancy group, in 2024.
This is the part that does not show up in a code-change bulletin: nobody flags a carried-forward field as wrong. It looks correct. It matches the last five projects. The general notes sheet is usually the least-scrutinized page in a QC pass precisely because it looks the most standardized. A reviewer checking dimension strings, door hardware sets, or life safety plan callouts has no natural reason to stop on a risk category designation that has not changed in years of practice, on the one year it did.
The distinction worth getting right before anyone asks
The other trap sits right next to the first one. A building designed for tornado loads under ASCE 7-22 Chapter 32 is not a storm shelter and does not meet ICC 500 or FEMA P-361. The design wind speeds under that provision for Risk Category III and IV work, roughly 80 to 125 mph and 95 to 140 mph respectively per the ICC writeup, are meant to reduce structural damage and improve life safety in ordinary construction. Storm shelters and safe rooms are a separate, much higher bar, designed for wind speeds around 250 mph in the central United States. A client who hears "this building is now designed for tornado loads" can reasonably assume they are getting a shelter. They are not, unless someone scoped ICC 500 separately. That is a conversation worth having in a proposal or an early code narrative, not in a change order after someone asks where the safe room is.
What this means for a QC checklist this year
Two checks, both cheap, worth adding to a firm's standard list this year:
First, confirm which code edition the jurisdiction of record has actually adopted, not which edition the team defaulted to from habit. Adoption timelines vary widely, and a new IBC edition can take up to about ten years to be adopted by a given state after ICC publishes it, per the same ICC technical writeup, so a firm working across several states can easily have projects on more than one edition at once.
Second, on any project with a qualifying Group I-2 component anywhere in the program, re-verify the risk category against the currently adopted table rather than the prior project's general notes sheet. If the jurisdiction has adopted the 2024 IBC and the site sits in the tornado-prone region, that same re-check determines whether the ASCE 7-22 tornado provision applies at all.
What this article does not tell you
It does not tell you whether your jurisdiction has adopted the 2024 IBC yet, or on what date, or with what local amendments. Adoption is set state by state and often city by city, grace periods vary, and Georgia's January 2026 effective date is one data point, not a pattern to extrapolate from. The only way to know is to check with the building department of record for the specific project. Everything above is a reason to check, not a substitute for checking.
Setmark's plan review work ingests a builder's checklist and drafts proposed rules for human review, the same re-derive-rather-than-assume habit this article is arguing for. That is not why this is worth knowing. It is worth knowing because the reclassification is real, it is already in effect somewhere, and it will not announce itself on the sheet.