ChatGPT UK Savings Rates Wrong Fscs 85000 Which

ChatGPT (OpenAI) (Free version of ChatGPT (model version not disclosed))FinanceOct 21

I asked ChatGPT: 'What's the best savings rate?' It began, without hesitation, with a table headed 'Typical UK savings rates (2026)' covering instant-access, one-year fixed-term savings (including Isas) and regular savers, without explaining what 'typical' meant or how the figures were derived, and it lumped fixed bonds together with fixed Isas. It quoted: instant-access average 2.15% AER; one-year fixed-term average 3.85% AER; one-year fixed-term cash Isa average 3.85% AER; regular savers up to 7.5% AER. For top rates it named 4% to 4.5% AER on instant-access accounts, up to 4.39% AER on instant-access cash Isas and 4.1% to 4.3% AER on one-year fixed-term cash Isas, and suggested no one-year fixed bond deals at all. Asked where to put a GBP 25,000 lump sum, it said the top 4.5% AER rate was available from Chase or Marcus by Goldman Sachs. Explaining why a cash Isa was 'great', it said savings were protected by the FSCS up to GBP 85,000, and repeated the same GBP 85,000 figure when asked directly how much was protected, producing the correct limit and a link to the FSCS site only after being asked to reveal its source. Is that right?

SHARE

1 Answer

0
✗ incorrectAI Corrector BotOct 21

Expert: Which? money team, UK consumer organisation, savings and consumer-affairs researchers Which? asked the free version of ChatGPT 'What's the best savings rate?' on 10 March 2026, then cross-referenced every figure it produced against Moneyfacts data for the same date. The result: of the four average rates the model quoted, exactly one was right. Average rates. Instant-access savings: ChatGPT said 2.15% AER, the actual average was 2.4%. One-year fixed-term savings: 3.85% against 3.79%. One-year fixed-term cash Isa: 3.85% against 3.76%. The only correct figure was regular savers, up to 7.5% AER. Top rates. Every 'top deal' it named was also inaccurate. Instant-access: 4% to 4.5% AER claimed, 5% AER available. Instant-access cash Isa: up to 4.39% claimed, 4.55% available. One-year fixed-term cash Isa: 4.1% to 4.3% claimed, 4.2% available. The model suggested no one-year fixed bond deals at all. Named providers. On a GBP 25,000 lump sum it told the reporter the top 4.5% AER rate was offered by Chase or by Marcus by Goldman Sachs. Chase does still pay 4.5% once its 2.25% bonus is counted, but Marcus was paying 3.75%. It also omitted two conditions that change the decision: Chase's rate drops to 2.25% when the bonus expires, and the product requires opening a current account first. (Which?'s own comparison put Cahoot's market-leading 5% AER ahead but capped at GBP 3,000 for 12 months.) A stale number that actually matters. ChatGPT said savings are protected by the Financial Services Compensation Scheme up to GBP 85,000. That limit was raised to GBP 120,000 per individual, per financial institution, on 1 December 2025, so the answer was months out of date on the one figure a saver needs if a bank fails. Asked directly how much was protected, it gave GBP 85,000 again; it corrected itself and linked to the FSCS only after the reporter demanded to see its source. It also assumed every provider is signed up to the scheme, when savers must check that themselves. The pattern behind it. Asked the same question on several separate occasions, the model gave different information each time, drawing on different sources, and it leaned on websites where rates and terms were out of date. Which?'s wider investigation found some chatbots, ChatGPT among them, gave advice that risked savers oversubscribing Isas in breach of HMRC rules. What it did get right. Its general strategy for a lump sum was sound: split the money rather than commit it to one account, fill the GBP 20,000 cash Isa allowance first, then place the rest in a top-rate instant-access account. It also correctly noted that the best account depends on what kind of savings product the customer wants, and that headline rates often come with withdrawal limits, temporary introductory rates or a current-account requirement. OpenAI's response to Which? was that its models are 'trained on data up to a certain point and responses do not incorporate information about events beyond that, unless tools are used', and that users researching consumer products should switch on the built-in web search tool, which shows and links its sources. It said its most-used model, GPT-5.3 Instant, gives more accurate answers when searching the web. Which?'s conclusion was that out-of-date rates and confidently wrong figures leave chatbots unsuitable for savings decisions: rates change weekly, and an answer without a live, checkable source is a guess dressed as advice. Source: https://www.which.co.uk/news/article/can-ai-really-find-you-the-best-savings-deal-aAy2E6f4Elky

Your answer

Sign in to verify this AI response.

Don't trust us — or the AI. Ask ChatGPT / Ask Claude / Ask Gemini this same question and compare the answers yourself.

More in Finance

More from this topic

Conversational AI chatbots1 answer

I asked a chatbot how to split my retirement savings between stocks and bonds. It answered in seconds, in tidy confident paragraphs, with no caveat and no follow-up question about my debts or my time horizon. I assumed it was right. Then Vanguard surveyed 6,686 of its own investors: of those who used AI for financial guidance, 37% said they got incorrect or misleading information, 44% said they could not tell whether it was right or wrong, and only 19% were confident the information was sound. Is that right?

Unidentified AI chatbots1 answer

I asked a chatbot to project my retirement from my superannuation balance. It came back with a complete, self-assured set of numbers. Then I read Deanne Stewart's column in the Australian Financial Review: when Aware Super's chief executive put superannuation questions to chatbots, one model invented details about the user's financial circumstances - including their salary - and hallucinated $100,000 of assets that did not exist, while another answered using outdated contribution cap limits. If the model is inventing the salary the whole projection is built on, what is that projection actually worth?

Coinbase AI news alert bot1 answer

I got a mass alert from Coinbase's automated breaking-news system stating that Norway's men's national team had clinched a quarter-final place at the FIFA World Cup by beating Brazil 3-2 - stated as settled fact. The alert went out before the match had kicked off. Norway did go on to win, but the final score was 2-1, so both the scoreline and the timing of the announcement were wrong. Did it get this wrong?