My new financial adviser is extremely helpful. She teaches me how to allocate my savings, build a house deposit and analyses if I have enough in my emergency fund.

But instead of discussing my money options in an office, my financial adviser is on a computer screen. In fact, she’s not real.

I’m talking to a chatbot driven by artificial intelligence. I decided to test it out for the month of January to see if it could jumpstart my New Year’s resolutions to rein in my spending – and start investing.

Of course, the gold standard of financial advice is finding a real, qualified, chartered financial adviser. However, this option is costly and inaccessible to many households.

And as interest in AI continues to soar, swathes of us will soon be asking it for help with our finances. Some 18million Britons have used AI for financial advice – or would consider doing so – says comparison website Finder.

However, as the information that AI generates is sometimes flawed and riddled with mistakes, I’m concerned about what financial information it might offer.

ChatGPT is one of the most popular generative AI tools. This means it uses algorithms to create content such as text or pictures.

Users can open the ChatGPT website – which you can either do on a smartphone or a computer – and type in a question, ask it to summarise information or ask it to write something, like a shopping list, for example.

Lucy Evans told the chatbot she wanted to revamp her finances this January

The chatbot uses information from books, news articles, journals and internet sources to write an original answer back in a conversational tone.

Before asking AI for help, I sat down with Tom Francis, head of digital advice at Octopus Money, to understand what good financial advice looks like.

He says: ‘You need someone to pick apart what the next 30 to 40 years might look like for you. That is easier with a human.

‘It might look like having the right amount in cash savings as an emergency fund or thinking about investing any spare savings in pensions and Isas. It needs to consider your future plans and if your finances are in a position to make investments.’

BOOSTING MY BUDGET

I start by telling the chatbot I want to revamp my finances this January. I enter my take home pay and expenses, plus how much I funnel into my saving pots each month.

The chatbot uses the 50-30-20 rule to tell me my fixed expenses are above what they should be (50 per cent of my take home pay), my ‘wants’ are correct (30 per cent), but I’m not saving as much as I should (20 per cent).

It tells me I should aim to boost my savings by just £50 a month if I can, which should be easy with the tips it gives me.

Meal prepping was high on the list – it even gave me suggestions of cheaper lunch ideas that range from £1 to £2.50 per portion.

For example, it suggests stir-fried rice, using leftover rice from dinner and any leftover veggies in the fridge, doused in soya sauce.

I estimate I saved around £45 this month by meal prepping for work lunches alone.

Another tip was to cancel any unused subscriptions. Britons spend an average of £117 a year on unused streaming subscriptions, says savings provider Raisin UK.

I check my direct debits and see I was paying for an Apple TV subscription which rolled over from a free trial. I cancel it, saving £8.99 each month from now on.

I also slash spending on takeaway coffees on the chatbot’s advice. While I cave a couple of times each week, I save £15 to £20 by not getting one every day.

In a bid to bolster my savings I also turn to the one part of my ‘fixed’ expenses I can cut – transport.

I spend around £200 a month on fares for the London Underground alone. Aside from paying rent on my property in south-west London, it’s my biggest fixed monthly expense.

AI says I should spend less on commuting and suggests working from home, carpooling, a bus pass or season tickets, none of which are suitable or available to me.

But it also suggests using an Oyster card instead of contactless payments. It tells me there’s a one-third discount on all Tube fares if I link my Railcard to my Oyster. I have a Railcard, bought for £30 for one year, that gives a third off rail travel for 16-25 year olds like me. I linked the two sometime last year so I dig it out from my purse at and top it up.

It says using my Oyster will save me some £11.20 a week on the commute from my home in zone three to our Kensington office in zone one.

The chatbot, however, does make a mistake – the one-third discount from linking the two cards only applies to off-peak fares. The chatbot wrongly tells me all pay-as-you-go fares will receive the one-third discount, no matter what time I travel.

You can link your Railcard to an Oyster at any Underground station – ask a clerk for help.

The chatbot also gives me cheaper restaurant options for supper with a group of friends. We settle on Franco Manca from a long list of options. I estimate my share of the bill was around £10 lower than what I’d typically pay for a meal out.

While the suggestions from AI were simple, it was refreshing to see how small cuts build up across several weeks. I easily saved £85 by meal prepping, buying fewer takeaway coffees, opting for a cheaper restaurant and cancelling a subscription.

Across 2025, that’s £1,020 which becomes £1,101 if left in a savings account paying 5 per cent for one year.

SAVVY SAVINGS ADVICE?

The AI adviser suggests using an Oyster card instead of contactless payments for travel

The AI adviser suggests using an Oyster card instead of contactless payments for travel

I must confess the rates on my savings accounts are not the top deals so ask the chatbot for help to boost my returns. I have an easy-access account which pays 3.1 per cent, a one-year fixed-rate bond at 3.9 per cent and a monthly saver paying 5.25 per cent.

The chatbot asks me a range of questions to find the right account for me, including if I am able to lock the money away, if I prefer traditional or online banks and if I need a branch account.

It eventually lands on a HSBC fixed-term savings account as one of the options, which pays 4.05 per cent fixed for one year.

I prefer traditional high street banks with branch and online accounts, but once I adjust my preference to include online banks, too, it recommends Marcus by Goldman Sachs. This online account pays 4.3 per cent.

While both options pay more than my current savings accounts, it disappoints me as I find neither option is top of the best buy tables.

I tell the chatbot my age and ask it how I should allocate my savings going forward and it tells me that 40 per cent of my savings should go towards retirement and just 20 per cent should go towards a house deposit, despite telling it this is my main financial goal. One recommendation is to funnel as much as 15 per cent of my salary into a pension.

While AI is right that boosting pension savings at a young age is vital to benefit from compound interest, I won’t hike my pension contributions to 15 per cent just yet.

Workers should typically halve their age to find out how much they should out aside between themselves and employer contributions. At age 23 my 6 per cent salary sacrifice, matched by my employer, is enough.

Mr Francis says to think about how this jump would impact my day-to-day spending and my tax band.

‘What if 6 per cent gets you to a retirement pot you are comfortable with? Financial planning is equally about saying how much more you can indulge, and spend on, as much as saying how much more to save.’

INVESTMENT ADVICE I WON’T TAKE

I also ask AI to talk me through my investment options and it tells me I may want a ‘mix of growth and safety’ in my portfolio.

It initially suggests I hold half of my portfolio in passive funds that are constructed using algorithms rather than by a fund manager – such as the Vanguard LifeStrategy 60 per cent Equity Fund.

It says on top of that, I should invest 30 per cent of my portfolio in bonds for steadier returns and the final 20 per cent should be invested in a mix of stocks if I am willing to take higher risk.

Mr Francis asks: ‘What has gone into this recommendation? What factors have fed into this? Has it considered cost, size, liquidity, performance or management style?’

I tell AI I have a low-risk appetite, favouring reliable and moderate growth, and it suggests a robo-adviser for a ‘hands-off, diversified investment with lower risk’.

Robo-advisers are automated platforms which manage investments with little to no human input.

It pushes the Nutmeg Conservative Portfolio, which it says invests 60 per cent of assets in bonds, which are typically more stable than stocks.

While this sounds impressive, I prefer to have more human interaction to help me get started.

Plus generative AI – which creates images and texts – creates content based on data it was trained with. This can be incorrect or biased.

Any money invested is at risk so it’s not a decision you want to get wrong.

I can’t trust that the chatbot properly considers my risk appetite, if my finances are in a position to invest or if this investment properly suits my goals. I decide to stick to my savings account and put all my efforts into reaching my goals for a house deposit.

Have you used AI for financial advice? Email L.evans@dailymail.co.uk

Some links in this article may be affiliate links. If you click on them we may earn a small commission. That helps us fund This Is Money, and keep it free to use. We do not write articles to promote products. We do not allow any commercial relationship to affect our editorial independence.

Share.
Exit mobile version