I asked DeepSeek and ChatGPT for financial advice on investing and retirement. Here's how the two rival chatbots stack up.
Here's how DeepSeek's R1 model stacks up against ChatGPT when it comes to giving financial advice.
admin
DeepSeek; ChatGPT; Getty Images; Chelsea Jia Feng/BI
- DeepSeek's R1 chatbot is the hottest new thing in the market and Apple's App Store.
- I decided to see how DeepSeek's low-cost AI model compares to ChatGPT in giving financial advice.
- While the chatbots differed in some aspects, I found the performance pretty comparable.
A few weeks ago, I asked ChatGPT for its 2025 market predictions.
Then, just last week, the Chinese AI startup DeepSeek released its newest R1 model, which turns out to be cheaper and more compute-efficient than OpenAI's ChatGPT.
I was curious to see if a competitor could deliver similar results from the same queries at a fraction of the cost and GPUs. I figured if DeepSeek's debut was impactful enough to wipe out over a trillion dollars in stock market value, including $589 billion of Nvidia's market cap, it probably has a pretty powerful product. So I asked DeepSeek and ChatGPT some of the same common personal finance questions.
ChatGPT debuted right as I finished college, meaning I narrowly missed being born in the generation using AI to cheat on — erm, I mean assist with — homework.
I was also probably born too early to experience a fully automated, robot-controlled society. But I was born just in time to ask two rival chatbots to give me some financial advice.
Here's how it went.
BI
This is a question I often hear, especially among my peer group of early 20-somethings new to investing.
R1 and ChatGPT gave me detailed, step-by-step guides that covered the basics, such as explaining investment terminology, choosing a type of investment account, diversifying between a mixture of stocks and bonds, and an example portfolio. Both chatbots suggested starting by dollar-cost averaging or investing a fixed amount at regular intervals.
While both chatbots covered similar content, I felt like R1 gave more concise and actionable recommendations. R1 suggested I start with index funds and gave several specific examples, while ChatGPT suggested a more open-ended combination of stocks, ETFs, and mutual funds.
There were also slight differences in the model portfolios. Both suggested putting 60% of one's money into US-based ETFs and stocks, but R1 suggested holding alternative investments such as gold or real estate as well. Meanwhile, ChatGPT did not include alternatives in its answer, instead suggesting a bigger bond allocation.
BI
An emergency fund covering three to six months of living expenses is an important step toward financial security, but saving up multiple thousands of dollars can be daunting.
I asked ChatGPT and R1 how they would tackle building an emergency fund, and their answers were quite similar. Both suggested starting small with a monthly savings goal and using automatic transfers to move the money into a dedicated savings account. R1 and ChatGPT gave specific actions to increase savings and cut spending, such as spending less on dining out, canceling unused subscriptions, and starting a side hustle.
One additional piece of advice that R1 gave was to track your progress with a chart or app to stay motivated.
BI
ChatGPT and R1 both told me that I would need around 70-80% of my pre-retirement income annually to maintain my lifestyle after retirement. That's a lot of money, and both chatbots were in agreement that there's no such thing as starting to save for retirement too early.
ChatGPT and R1 suggested taking advantage of employer-sponsored plans such as a 401(k), and opening individual retirement accounts like a Roth IRA or traditional IRA.
ChatGPT gave additional suggestions, such as using a health savings account or a target-date fund that automatically adjusts its stock and bond allocation as you approach retirement.
ChatGPT also cautioned against taking on too much risk later in life.
"As you get closer to retirement, you may want to gradually reduce risk by shifting towards more conservative investments (e.g., bonds or cash equivalents) to protect your savings," it told me. R1 didn't provide such a caveat — its recommendations seemed more tailored toward a young adult's short-term investing goals.
This can be seen in the specific retirement plans R1 and ChatGPT provided me. ChatGPT was more cognizant of dialing down the risk at age 40 and beyond, while R1 didn't mention switching up the retirement portfolio allocation later in life. R1's 90/10 stock-to-bond allocation might be suitable for a 25-year-old but probably not for a 40-year-old.
Anthony Kwan/Getty Images
I found both DeepSeek and OpenAI's models to be pretty comparable when it came to financial advice. That's a big deal, considering that DeepSeek's offering costs 20 to 40 times less to produce than OpenAI's.
There were times when one chatbot provided a new suggestion not covered by the other, but most of the responses were similar and in line with what a financial advisor would suggest. The questions I asked the chatbots were also pretty open-ended, so I'm sure a more detailed prompt would yield more specific suggestions.
Unlike some products, I don't think there are significant switching costs for one chatbot versus another. I would be hard-pressed to switch from Spotify to Apple Music, for example, but that's not the case for R1 and ChatGPT, as both are extremely simple to set up and use.
That being said, nobody should make multiple-thousand-dollar decisions solely based on the advice of a chatbot. But they are useful tools for researching complicated topics like personal finance.
