We Asked 7 Versions of GPT to Pick
Australia's Top Growth Suburbs.
Then We Checked.

1. The One-Line Summary

2. How We Tested This

Every version of GPT has a knowledge cutoff date. GPT-3.5 stopped learning in late 2021. GPT-4o stopped in late 2023. That creates a natural experiment. We can ask each version to forecast property growth, then check the actual prices for the years that followed.

We queried 7 versions of GPT across all 8 Australian states and territories. Each version was asked to name 20 suburbs most likely to grow, along with expected growth percentages and reasoning. We then measured actual suburb growth from the month after each version's exact publication date through to February 2026. We used monthly hedonic price data.

The critical part is simple. We compare each version's picks against all other suburbs in the same state, over the same period. This controls for the broader market. If Sydney grew 15% overall, we are not asking whether GPT predicted 15%. We are asking whether the specific suburbs it chose grew more or less than 15%.

GPT Versions Tested

3. Price Growth: GPT Picks vs Market

These charts show indexed price growth from each version's publication date. Both lines start at 100. The solid line tracks the average hedonic price of GPT-picked suburbs. The dashed line shows the market baseline. If the solid line ends above the dashed line, GPT beat the market.

4. Did GPT Pick Winners?

This is the only question that matters for investors. If you followed GPT's suburb recommendations, did your property grow faster than average? The answer, for most versions, is no. Dollar costs are calculated at the median property price of each model's picks.

5. Performance by State

GPT's biggest miss was Western Australia. Across all GPT versions combined, WA picks underperformed by 8.67 percentage points. That cost investors $84,028 on the median $969,000 WA property. Only 24% of GPT's WA picks beat the state median.

Victoria was the only state where GPT picks slightly outperformed, gaining $5,652 on the median $803,000 property. But 54% beating the median is barely above random chance.

6. The Prompt We Used

We sent the same prompt to every version, for every state and territory in Australia:

What GPT-3.5 Said (NSW, up to 2024)

Notice the generic reasoning. "Infrastructure development", "proximity to Sydney", "strong demand". These phrases appear thousands of times in real estate articles. GPT is reciting training text patterns, not analysing live market signals.

7. Real Properties. Real Losses.

These are not hypotheticals. GPT recommended these suburbs, some as top picks. Anyone who bought where GPT told them to lost money, before transaction costs.

GPT's Worst Advice

Even a Broken Clock Can Be Right

The wins were accidental. In every case GPT predicted modest single-digit growth (3-8%). It named the suburbs but had no idea they would boom. Meanwhile, its #2 picks included Sandy Bay (now $1,140,000, down 15%) and Launceston (now $817,000, down 13.4%). You can verify these prices on our suburb report pages.

8. Absolute Accuracy: How Far Off Were the Numbers?

Beyond picking the right suburbs, how accurate were the growth percentage predictions themselves? Every version under-predicted actual growth except gpt-4.1-mini. The Australian market grew far faster than any version of GPT expected.

9. Data Validation: Microburbs vs CoreLogic and Domain

We cross-checked our Microburbs hedonic price data against published figures from Cotality (formerly CoreLogic) and Domain. The state rankings match exactly. Growth magnitudes are directionally aligned, with differences explained by methodology.

10. What This Means for Property Investors

GPT is trained on internet text. It knows which suburbs get talked about. It knows the cliches: "close to the CBD", "infrastructure investment", "lifestyle appeal". But popular suburbs are already priced in. The growth happens in the places nobody is talking about yet.

11. Methodology

12. Conclusion

You would have been better off picking suburbs at random.

Across 6,888 scored predictions from 7 GPT versions, six underperformed the market. The only version that beat the market (GPT-4.1, 1,048 picks) gained $3,599 on an $861,000 property over 9 months. That is noise, not signal.

The worst performers cost real money. GPT-4.1-mini cost investors $26,729 on an $884,000 property. GPT-4-turbo cost $15,945 on a $950,000 property. GPT-3.5 cost $14,994 on an $882,000 property. In Western Australia alone, GPT cost investors $84,028 on the median property.

Property forecasting requires live, hyper-local data. Supply pipelines, demographic flows, approval volumes, wage growth by postcode. These signals exist, and they work. But they live in databases, not in the training text of a language model.

Do not ask ChatGPT where to buy property. A random selection of suburbs would have outperformed GPT in six of seven cases.