ChatGPT as a Financial Advisor: A Re-Examination
Bibliographic record
Abstract
Building on prior research, we revisited the 21 personal finance scenarios using OpenAI’s newer ChatGPT-4o to observe whether its financial guidance has meaningfully evolved. Our qualitative analysis relied on expert assessments to examine both the content and tone of the model’s advice, considering how prompt engineering influenced ChatGPT outputs. We observed that ChatGPT-4o often produced more thorough suggestions and paid closer attention to tax implications—though it still overlooked some important details. It also showed more creative thinking in certain situations. However, some of the same shortcomings persisted: Generalizations remained too broad with respect to certain topics, legal references were occasionally misleading, and emotional empathy continued to feel artificial, even with carefully crafted prompts. We also extended our analysis to the newest ChatGPT model (ChatGPT-5). We found that the recommendations generated by ChatGPT-5 were quite similar to those generated by ChatGPT-4o, but the accuracy in the numerical problems was better under ChatGPT-5. While not a replacement for financial professionals, ChatGPT appears to be maturing into a more useful supporting tool for both advisors and clients. Our findings not only suggest cautious optimism but also underscore the need for careful oversight when using such tools in personal financial decision-making.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.055 | 0.221 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.004 | 0.006 |
| Scholarly communication | 0.011 | 0.020 |
| Open science | 0.004 | 0.006 |
| Research integrity | 0.003 | 0.007 |
| Insufficient payload (model declined to judge) | 0.008 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".