AI chatbots give wrong answers to financial queries 'most of the time'

11vuio0pswjnm7 about 2 hours ago 19 commentsRead Article on ft.com

ES version is available. Content is displayed in original English for accuracy.

⚡ Community Insights

Discussion Sentiment

71% Positive

Analyzed from 668 words in the discussion.

Discussion (19 Comments)Read Original on HackerNews

lukeify•24 minutes ago

Given most financial advisors tend to vend out suboptimal advice and steer customers in favour of products they receive a kickback for, I'm happy to be accepting of an unbiased LLM that's trained on bogleheads.org.

in_absentia•16 minutes ago

If that's what you want, I'll save you some tokens:

#!/bin/sh

while read question; do echo "Put it into VFIAX"; done

ehe78qhe•6 minutes ago

This is missing a lot of steps like:

- Building an emergency fund

- Budgeting and tracking where your money goes

- Planning and saving for large purchases like cars, homes and life goals

- Optimizing use of tax-advantaged accounts like 401Ks, HSAs, and IRAs

- What to do with ESPPs, RSUs, and options

- How taxes work and how to optimize around them

- Estate planning

ehe78qhe•about 1 hour ago

Article is just a vague summary of https://www.saturnos.com/report/artificial-authority

Anecdotally, current models seem to be decent at general personal finance principles - certainly better than the majority of personal finance education that people get exposed to unless they seek it out and read a variety of books and sources. But I wouldn't trust them with direct decision making with actual money due to the training lag time on current tax policy, etc.

NitpickLawyer•27 minutes ago

> certainly better than the majority of personal finance education that people get exposed to unless they seek it out and read a variety of books and sources.

Also, models are now good enough that you can give them chapters from "authoritative" books, and they'll integrate that and come up with better answers even if their "vanilla" answers were average. And they'll tailor stuff to your particular situation. It's funny that the "agentic" stuff is only used in coding mostly, while it can and does work in other fields as well.

As always, you kinda need to check it (at least spot check) but all in all I'd agree it's better than the average stuff you used to find with a quick google search.

bluecalm•2 minutes ago

So I downloaded that report which of course doesn't contain the most relevant information (the questions) but it contains some examples of wrong answers.

I fed the first question to Grok (which they claimed they tested as well) and it answered it correctly in detail.

I repeated it with another one - again correct answer. I then selected the question they said Grok specifically answered incorrectly and it again answered it correctly.

I am sticking with my first intuition: journalists are terrible at testing tools and probably wanted them to answer incorrectly/not fully (the questions are constructed in a way to make it difficult as well).

People reading ft will now think chat boxes are bad at answering financial questions while they are pretty good at it. Zero consequences for spreading fake news for Financial Times there but good for financial advisors I guess.

01100011•about 1 hour ago

Single shot or with reasoning enabled? My experience is that reasoning dramatically reduces hallucinations and improves output quality. I don't trust models without it.

wonnage•13 minutes ago

> Overall, the best-performing model was Claude Opus 5 on “reasoning” mode, which still made mistakes in 39 per cent of answers.

in_absentia•35 minutes ago

Now, compare this to a recent story that seemed to claim the opposite:

https://news.ycombinator.com/item?id=49139102

I don't have the time to review the underlying research and decide which one is more correct. My personal biases make me want to believe the current one. Your personal biases may be pulling you in the other direction. How do we make the conversation more intelligent than that?

sixtyj•about 1 hour ago

I would prefer to use agent-assisted python scripts that chatbot.

k7peak•about 1 hour ago

Agreed, this works really well for me. Double check the math/python, execute many times without a LLM that can change o

yieldcrv•9 minutes ago

Tool calling solves this

74gee•28 minutes ago

Well duh! If it's not using tools to look up the state of the market empirically it's not likely to be accurate financially.

demibabs•about 1 hour ago

I like how FT makes me accept cookies from their 46 “technology” (advertising) partners before showing me that the article is behind a paywall anyway.

jb1991•about 1 hour ago

You actually like that? I find it kind of annoying.

Wololooo•about 1 hour ago

No they do not like it, it is a figure of speech to underline how much they do not like it.

simianwords•44 minutes ago

These models do pretty well in benchmarks and real world so I'm highly suspicious of this article. Further more, in the original report, the examples of bad answers are from Haiku - at least 7 out of 10. Anyone who knows anything about LLMs know that haiku shouldn't be used for anything pretty much.

There's no reproducible set either. I'm not gonna trust this report.

stymaar•7 minutes ago

Most people[1] interacting with chatbots don't have a paid subscription and they do interact with the free-tier LLMs that are Luna and Haiku, so I still think it's relevant.

[1]: not on HN obviously, but IRL, and probably among FT's readership as well.

SyneRyder•19 minutes ago

Another summary report here, for those who can't get past the paywall:

https://www.financialreporter.co.uk/ai-models-give-wrong-fin...

Much of the testing is on Haiku and Luna, and criticizing the quality of free AI (!). But they do claim Opus 5 with reasoning still failed 39% of their financial questions.